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 | 22.174 ms | 0 - 43 MB | NPU | -- |
| Yolo-v6 | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 14.686 ms | 1 - 11 MB | NPU | -- |
| Yolo-v6 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 12.481 ms | 0 - 42 MB | NPU | -- |
| Yolo-v6 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 7.335 ms | 5 - 41 MB | NPU | -- |
| Yolo-v6 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 10.773 ms | 0 - 16 MB | NPU | -- |
| Yolo-v6 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 4.485 ms | 5 - 13 MB | NPU | -- |
| Yolo-v6 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 12.586 ms | 0 - 43 MB | NPU | -- |
| Yolo-v6 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 6.204 ms | 1 - 13 MB | NPU | -- |
| Yolo-v6 | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 22.174 ms | 0 - 43 MB | NPU | -- |
| Yolo-v6 | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 14.686 ms | 1 - 11 MB | NPU | -- |
| Yolo-v6 | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 10.867 ms | 0 - 15 MB | NPU | -- |
| Yolo-v6 | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 4.53 ms | 5 - 7 MB | NPU | -- |
| Yolo-v6 | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 13.59 ms | 0 - 30 MB | NPU | -- |
| Yolo-v6 | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 6.974 ms | 0 - 17 MB | NPU | -- |
| Yolo-v6 | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 10.76 ms | 0 - 15 MB | NPU | -- |
| Yolo-v6 | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 4.596 ms | 5 - 7 MB | NPU | -- |
| Yolo-v6 | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 12.586 ms | 0 - 43 MB | NPU | -- |
| Yolo-v6 | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 6.204 ms | 1 - 13 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 10.765 ms | 0 - 16 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 4.498 ms | 0 - 45 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 5.57 ms | 1 - 38 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 7.437 ms | 0 - 59 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 3.186 ms | 5 - 140 MB | NPU | -- |
| Yolo-v6 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 4.106 ms | 0 - 158 MB | NPU | -- |
| Yolo-v6 | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 7.537 ms | 0 - 50 MB | NPU | -- |
| Yolo-v6 | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 2.52 ms | 5 - 76 MB | NPU | -- |
| Yolo-v6 | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 3.521 ms | 5 - 103 MB | NPU | -- |
| Yolo-v6 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 4.758 ms | 5 - 5 MB | NPU | -- |
| Yolo-v6 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.048 ms | 7 - 7 MB | NPU | -- |
| Yolo-v6 | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 5.191 ms | 1 - 11 MB | NPU | -- |
| Yolo-v6 | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 2.738 ms | 2 - 36 MB | NPU | -- |
| Yolo-v6 | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 2.166 ms | 2 - 11 MB | NPU | -- |
| Yolo-v6 | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 2.757 ms | 1 - 12 MB | NPU | -- |
| Yolo-v6 | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN | 7.814 ms | 2 - 17 MB | NPU | -- |
| Yolo-v6 | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN | 5.191 ms | 1 - 11 MB | NPU | -- |
| Yolo-v6 | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 2.169 ms | 3 - 5 MB | NPU | -- |
| Yolo-v6 | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN | 3.343 ms | 2 - 18 MB | NPU | -- |
| Yolo-v6 | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 2.209 ms | 4 - 5 MB | NPU | -- |
| Yolo-v6 | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN | 2.757 ms | 1 - 12 MB | NPU | -- |
| Yolo-v6 | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 2.166 ms | 2 - 11 MB | NPU | -- |
| Yolo-v6 | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 5.612 ms | 0 - 30 MB | NPU | -- |
| Yolo-v6 | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 1.483 ms | 2 - 42 MB | NPU | -- |
| Yolo-v6 | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 3.844 ms | 3 - 103 MB | NPU | -- |
| Yolo-v6 | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 1.398 ms | 2 - 35 MB | NPU | -- |
| Yolo-v6 | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 3.247 ms | 1 - 100 MB | NPU | -- |
| Yolo-v6 | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.472 ms | 2 - 2 MB | NPU | -- |
| Yolo-v6 | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 6.33 ms | 3 - 3 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 4.482 ms | 0 - 20 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 3.161 ms | 1 - 11 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 2.258 ms | 0 - 39 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 1.656 ms | 1 - 31 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 2.063 ms | 0 - 31 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 1.342 ms | 1 - 5 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 2.511 ms | 0 - 21 MB | NPU | -- |
| Yolo-v6 | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 1.662 ms | 1 - 13 MB | NPU | -- |
| Yolo-v6 | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | TFLITE | 4.648 ms | 0 - 28 MB | NPU | -- |
| Yolo-v6 | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN | 4.854 ms | 1 - 13 MB | NPU | -- |
| Yolo-v6 | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 35.771 ms | 3 - 12 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 4.482 ms | 0 - 20 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA7255P ADP | Qualcomm® SA7255P | QNN | 3.161 ms | 1 - 11 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 2.078 ms | 0 - 30 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 1.346 ms | 1 - 4 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 2.995 ms | 0 - 26 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8295P ADP | Qualcomm® SA8295P | QNN | 2.275 ms | 0 - 18 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 2.074 ms | 0 - 29 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 1.34 ms | 0 - 2 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 2.511 ms | 0 - 21 MB | NPU | -- |
| Yolo-v6 | w8a8 | SA8775P ADP | Qualcomm® SA8775P | QNN | 1.662 ms | 1 - 13 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 2.077 ms | 0 - 30 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 1.368 ms | 1 - 27 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 3.537 ms | 0 - 34 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 1.351 ms | 0 - 41 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 0.905 ms | 1 - 29 MB | NPU | -- |
| Yolo-v6 | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 2.432 ms | 0 - 116 MB | NPU | -- |
| Yolo-v6 | w8a8 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 1.104 ms | 0 - 25 MB | NPU | -- |
| Yolo-v6 | w8a8 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN | 0.814 ms | 1 - 25 MB | NPU | -- |
| Yolo-v6 | w8a8 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 2.593 ms | 1 - 98 MB | NPU | -- |
| Yolo-v6 | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 1.517 ms | 1 - 1 MB | NPU | -- |
| Yolo-v6 | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 3.83 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