YOLOv11-Detection-Quantized: Optimized for Mobile Deployment
Quantized real-time object detection optimized for mobile and edge by Ultralytics
Ultralytics YOLOv11 is a machine learning model that predicts bounding boxes and classes of objects in an image. This model is post-training quantized to int8 using samples from the COCO dataset.
This model is an implementation of YOLOv11-Detection-Quantized found here.
More details on model performance across various devices, can be found here.
Model Details
- Model Type: Object detection
- Model Stats:
- Model checkpoint: YOLOv11-N
- Input resolution: 640x640
- Number of parameters: 2.64M
- Model size: 2.83 MB
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model |
|---|---|---|---|---|---|---|---|---|
| YOLOv11-Detection-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 1.833 ms | 0 - 12 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 1.822 ms | 1 - 4 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 10.152 ms | 0 - 20 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.202 ms | 0 - 34 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 1.223 ms | 1 - 20 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 7.445 ms | 1 - 64 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 0.924 ms | 0 - 33 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.24 ms | 1 - 30 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 7.109 ms | 2 - 63 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA7255P ADP | SA7255P | TFLITE | 9.079 ms | 0 - 22 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA7255P ADP | SA7255P | QNN | 8.966 ms | 1 - 9 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.832 ms | 0 - 13 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8255 (Proxy) | SA8255P Proxy | QNN | 1.817 ms | 1 - 4 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8295P ADP | SA8295P | TFLITE | 2.662 ms | 0 - 26 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8295P ADP | SA8295P | QNN | 2.607 ms | 1 - 16 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.819 ms | 0 - 10 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8650 (Proxy) | SA8650P Proxy | QNN | 1.823 ms | 1 - 4 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8775P ADP | SA8775P | TFLITE | 2.715 ms | 0 - 22 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | SA8775P ADP | SA8775P | QNN | 2.688 ms | 1 - 11 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 4.081 ms | 0 - 28 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 5.937 ms | 1 - 13 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 66.03 ms | 1 - 11 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 9.079 ms | 0 - 22 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 8.966 ms | 1 - 9 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.821 ms | 0 - 8 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 1.804 ms | 1 - 4 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 2.715 ms | 0 - 22 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 2.688 ms | 1 - 11 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 1.98 ms | 0 - 30 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 2.242 ms | 1 - 30 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.097 ms | 1 - 1 MB | INT8 | NPU | -- |
| YOLOv11-Detection-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 10.967 ms | 2 - 2 MB | INT8 | NPU | -- |
License
- The license for the original implementation of YOLOv11-Detection-Quantized 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
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The HF Inference API does not support object-detection models for pytorch library.