Yolo-v6: Optimized for Mobile Deployment
Real-time object detection optimized for mobile and edge
YoloV6 is a machine learning model that predicts bounding boxes and classes of objects in an image.
This model is an implementation of Yolo-v6 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: YoloV6-N
- Input resolution: 640x640
- Number of parameters: 4.68M
- Model size: 17.9 MB
| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model |
|---|---|---|---|---|---|---|---|---|
| Yolo-v6 | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 21.725 ms | 0 - 38 MB | NPU | -- |
| Yolo-v6 | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 14.794 ms | 1 - 11 MB | NPU | -- |
| Yolo-v6 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 12.844 ms | 0 - 40 MB | NPU | -- |
| Yolo-v6 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 7.697 ms | 5 - 39 MB | NPU | -- |
| Yolo-v6 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 10.512 ms | 0 - 13 MB | NPU | -- |
| Yolo-v6 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 4.493 ms | 5 - 8 MB | NPU | -- |
| Yolo-v6 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 12.4 ms | 0 - 39 MB | NPU | -- |
| Yolo-v6 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 6.315 ms | 1 - 16 MB | NPU | -- |
| Yolo-v6 | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 21.725 ms | 0 - 38 MB | NPU | -- |
| Yolo-v6 | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 14.794 ms | 1 - 11 MB | NPU | -- |
| Yolo-v6 | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 10.579 ms | 0 - 11 MB | NPU | -- |
| Yolo-v6 | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 4.748 ms | 5 - 7 MB | NPU | -- |
| Yolo-v6 | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 13.897 ms | 0 - 29 MB | NPU | -- |
| Yolo-v6 | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 7.072 ms | 0 - 18 MB | NPU | -- |
| Yolo-v6 | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 10.621 ms | 0 - 8 MB | NPU | -- |
| Yolo-v6 | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 4.482 ms | 5 - 8 MB | NPU | -- |
| Yolo-v6 | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 12.4 ms | 0 - 39 MB | NPU | -- |
| Yolo-v6 | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 6.315 ms | 1 - 16 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 10.614 ms | 0 - 11 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 4.495 ms | 5 - 17 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 5.446 ms | 1 - 35 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 7.263 ms | 0 - 49 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 3.192 ms | 0 - 49 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 4.208 ms | 7 - 73 MB | NPU | -- |
| Yolo-v6 | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 7.201 ms | 0 - 43 MB | NPU | -- |
| Yolo-v6 | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 2.471 ms | 5 - 45 MB | NPU | -- |
| Yolo-v6 | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 4.264 ms | 5 - 55 MB | NPU | -- |
| Yolo-v6 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 4.775 ms | 5 - 5 MB | NPU | -- |
| Yolo-v6 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.064 ms | 6 - 6 MB | NPU | -- |
| Yolo-v6 | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 4.851 ms | 0 - 10 MB | NPU | -- |
| Yolo-v6 | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 2.615 ms | 2 - 38 MB | NPU | -- |
| Yolo-v6 | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 1.893 ms | 2 - 5 MB | NPU | -- |
| Yolo-v6 | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 2.488 ms | 2 - 16 MB | NPU | -- |
| Yolo-v6 | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN | 8.243 ms | 2 - 13 MB | NPU | -- |
| Yolo-v6 | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN | 4.851 ms | 0 - 10 MB | NPU | -- |
| Yolo-v6 | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 1.907 ms | 2 - 4 MB | NPU | -- |
| Yolo-v6 | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN | 2.974 ms | 0 - 18 MB | NPU | -- |
| Yolo-v6 | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 1.89 ms | 2 - 5 MB | NPU | -- |
| Yolo-v6 | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN | 2.488 ms | 2 - 16 MB | NPU | -- |
| Yolo-v6 | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 1.945 ms | 2 - 11 MB | NPU | -- |
| Yolo-v6 | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 5.395 ms | 0 - 33 MB | NPU | -- |
| Yolo-v6 | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 1.311 ms | 2 - 37 MB | NPU | -- |
| Yolo-v6 | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 3.723 ms | 4 - 76 MB | NPU | -- |
| Yolo-v6 | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 1.214 ms | 2 - 30 MB | NPU | -- |
| Yolo-v6 | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 3.276 ms | 1 - 67 MB | NPU | -- |
| Yolo-v6 | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.163 ms | 2 - 2 MB | NPU | -- |
| Yolo-v6 | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.252 ms | 3 - 3 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 4.484 ms | 0 - 20 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 3.143 ms | 1 - 11 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 2.4 ms | 0 - 34 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 1.857 ms | 1 - 36 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 2.081 ms | 0 - 27 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 1.347 ms | 1 - 4 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 2.47 ms | 0 - 22 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 1.674 ms | 1 - 15 MB | NPU | -- |
| Yolo-v6 | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | TFLITE | 4.464 ms | 0 - 31 MB | NPU | -- |
| Yolo-v6 | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN | 4.586 ms | 1 - 16 MB | NPU | -- |
| Yolo-v6 | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 33.498 ms | 3 - 12 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 4.484 ms | 0 - 20 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA7255P ADP | Qualcomm® SA7255P | QNN | 3.143 ms | 1 - 11 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 2.073 ms | 0 - 27 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 1.348 ms | 1 - 12 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 2.956 ms | 0 - 22 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8295P ADP | Qualcomm® SA8295P | QNN | 2.357 ms | 1 - 19 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 2.061 ms | 0 - 27 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 1.359 ms | 1 - 4 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 2.47 ms | 0 - 22 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8775P ADP | Qualcomm® SA8775P | QNN | 1.674 ms | 1 - 15 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 2.079 ms | 0 - 28 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 1.364 ms | 0 - 25 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 3.17 ms | 0 - 33 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 1.389 ms | 0 - 32 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 0.907 ms | 1 - 35 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 2.284 ms | 0 - 68 MB | NPU | -- |
| Yolo-v6 | w8a8 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 1.331 ms | 0 - 25 MB | NPU | -- |
| Yolo-v6 | w8a8 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 0.663 ms | 1 - 33 MB | NPU | -- |
| Yolo-v6 | w8a8 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 2.511 ms | 1 - 58 MB | NPU | -- |
| Yolo-v6 | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 1.519 ms | 1 - 1 MB | NPU | -- |
| Yolo-v6 | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 3.638 ms | 4 - 4 MB | NPU | -- |
License
- The license for the original implementation of Yolo-v6 can be found here.
- The license for the compiled assets for on-device deployment can be found here
References
- YOLOv6: A Single-Stage Object Detection Framework for Industrial Applications
- Source Model Implementation
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
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
🙋
Ask for provider support