metadata
library_name: pytorch
license: other
tags:
- real_time
- android
pipeline_tag: object-detection
YOLOv8-Detection: Optimized for Mobile Deployment
Real-time object detection optimized for mobile and edge by Ultralytics
Ultralytics YOLOv8 is a machine learning model that predicts bounding boxes and classes of objects in an image.
This model is an implementation of YOLOv8-Detection found here.
More details on model performance across various devices, can be found here.
Model Details
- Model Type: Model_use_case.object_detection
- Model Stats:
- Model checkpoint: YOLOv8-N
- Input resolution: 640x640
- Number of parameters: 3.18M
- Model size (float): 12.2 MB
- Model size (w8a8): 3.26 MB
| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model |
|---|---|---|---|---|---|---|---|---|
| YOLOv8-Detection | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 16.799 ms | 0 - 28 MB | NPU | -- |
| YOLOv8-Detection | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 13.523 ms | 1 - 10 MB | NPU | -- |
| YOLOv8-Detection | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 8.885 ms | 0 - 37 MB | NPU | -- |
| YOLOv8-Detection | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 6.976 ms | 5 - 33 MB | NPU | -- |
| YOLOv8-Detection | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 5.861 ms | 0 - 15 MB | NPU | -- |
| YOLOv8-Detection | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 4.056 ms | 5 - 7 MB | NPU | -- |
| YOLOv8-Detection | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 7.285 ms | 0 - 29 MB | NPU | -- |
| YOLOv8-Detection | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 5.533 ms | 2 - 16 MB | NPU | -- |
| YOLOv8-Detection | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 16.799 ms | 0 - 28 MB | NPU | -- |
| YOLOv8-Detection | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 13.523 ms | 1 - 10 MB | NPU | -- |
| YOLOv8-Detection | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 5.812 ms | 0 - 17 MB | NPU | -- |
| YOLOv8-Detection | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 4.091 ms | 5 - 7 MB | NPU | -- |
| YOLOv8-Detection | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 9.577 ms | 0 - 25 MB | NPU | -- |
| YOLOv8-Detection | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 7.733 ms | 0 - 18 MB | NPU | -- |
| YOLOv8-Detection | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 5.764 ms | 0 - 13 MB | NPU | -- |
| YOLOv8-Detection | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 4.053 ms | 5 - 7 MB | NPU | -- |
| YOLOv8-Detection | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 7.285 ms | 0 - 29 MB | NPU | -- |
| YOLOv8-Detection | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 5.533 ms | 2 - 16 MB | NPU | -- |
| YOLOv8-Detection | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 5.835 ms | 0 - 17 MB | NPU | -- |
| YOLOv8-Detection | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 4.008 ms | 6 - 18 MB | NPU | -- |
| YOLOv8-Detection | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 5.382 ms | 5 - 30 MB | NPU | -- |
| YOLOv8-Detection | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 4.194 ms | 0 - 38 MB | NPU | -- |
| YOLOv8-Detection | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 2.803 ms | 0 - 42 MB | NPU | -- |
| YOLOv8-Detection | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 3.6 ms | 1 - 50 MB | NPU | -- |
| YOLOv8-Detection | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 4.057 ms | 0 - 33 MB | NPU | -- |
| YOLOv8-Detection | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 2.609 ms | 5 - 38 MB | NPU | -- |
| YOLOv8-Detection | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 3.661 ms | 5 - 42 MB | NPU | -- |
| YOLOv8-Detection | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 4.468 ms | 5 - 5 MB | NPU | -- |
| YOLOv8-Detection | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.029 ms | 5 - 5 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 7.681 ms | 2 - 12 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 4.893 ms | 2 - 41 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 3.958 ms | 2 - 5 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 4.546 ms | 2 - 16 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN | 15.377 ms | 2 - 18 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN | 7.681 ms | 2 - 12 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 3.934 ms | 2 - 4 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN | 5.194 ms | 0 - 18 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 3.937 ms | 6 - 8 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN | 4.546 ms | 2 - 16 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 3.944 ms | 2 - 14 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 9.216 ms | 3 - 23 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 2.606 ms | 2 - 45 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 6.718 ms | 0 - 47 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 2.267 ms | 2 - 39 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 5.431 ms | 2 - 47 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 4.347 ms | 2 - 2 MB | NPU | -- |
| YOLOv8-Detection | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 9.609 ms | 1 - 1 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 3.356 ms | 0 - 20 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 3.348 ms | 1 - 11 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 1.828 ms | 1 - 31 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 2.033 ms | 1 - 33 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 1.515 ms | 0 - 13 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 1.574 ms | 1 - 3 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 1.915 ms | 0 - 22 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 1.963 ms | 1 - 16 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | TFLITE | 3.7 ms | 0 - 30 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN | 5.767 ms | 1 - 15 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 45.826 ms | 3 - 12 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 3.356 ms | 0 - 20 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | SA7255P ADP | Qualcomm® SA7255P | QNN | 3.348 ms | 1 - 11 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 1.505 ms | 0 - 12 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 1.576 ms | 1 - 3 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 2.363 ms | 0 - 23 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | SA8295P ADP | Qualcomm® SA8295P | QNN | 2.336 ms | 0 - 20 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 1.524 ms | 0 - 13 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 1.579 ms | 1 - 3 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 1.915 ms | 0 - 22 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | SA8775P ADP | Qualcomm® SA8775P | QNN | 1.963 ms | 1 - 16 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 1.503 ms | 0 - 12 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 1.592 ms | 1 - 11 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 5.848 ms | 0 - 17 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 1.0 ms | 0 - 30 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 1.04 ms | 1 - 35 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 4.159 ms | 0 - 47 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 0.771 ms | 1 - 29 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 0.896 ms | 1 - 30 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 3.661 ms | 1 - 41 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 1.813 ms | 1 - 1 MB | NPU | -- |
| YOLOv8-Detection | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.58 ms | 0 - 0 MB | NPU | -- |
License
- The license for the original implementation of YOLOv8-Detection can be found here.
- The license for the compiled assets for on-device deployment can be found here
References
Community
- Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
- For questions or feedback please reach out to us.
Usage and Limitations
Model may not be used for or in connection with any of the following applications:
- Accessing essential private and public services and benefits;
- Administration of justice and democratic processes;
- Assessing or recognizing the emotional state of a person;
- Biometric and biometrics-based systems, including categorization of persons based on sensitive characteristics;
- Education and vocational training;
- Employment and workers management;
- Exploitation of the vulnerabilities of persons resulting in harmful behavior;
- General purpose social scoring;
- Law enforcement;
- Management and operation of critical infrastructure;
- Migration, asylum and border control management;
- Predictive policing;
- Real-time remote biometric identification in public spaces;
- Recommender systems of social media platforms;
- Scraping of facial images (from the internet or otherwise); and/or
- Subliminal manipulation