MobileNet-v2: Optimized for Qualcomm Devices
MobileNetV2 is a machine learning model that can classify images from the Imagenet dataset. It can also be used as a backbone in building more complex models for specific use cases.
This is based on the implementation of MobileNet-v2 found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.
Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.
Getting Started
There are two ways to deploy this model on your device:
Option 1: Download Pre-Exported Models
Below are pre-exported model assets ready for deployment.
| Runtime | Precision | Chipset | SDK Versions | Download |
|---|---|---|---|---|
| ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | Download |
| ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | Download |
| ONNX | w8a16_mixed_int16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | Download |
| ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | Download |
| QNN_DLC | float | Universal | QAIRT 2.45 | Download |
| QNN_DLC | w8a16 | Universal | QAIRT 2.45 | Download |
| QNN_DLC | w8a16_mixed_int16 | Universal | QAIRT 2.45 | Download |
| QNN_DLC | w8a8 | Universal | QAIRT 2.45 | Download |
| TFLITE | float | Universal | QAIRT 2.45 | Download |
| TFLITE | w8a8 | Universal | QAIRT 2.45 | Download |
For more device-specific assets and performance metrics, visit MobileNet-v2 on Qualcomm® AI Hub.
Option 2: Export with Custom Configurations
Use the Qualcomm® AI Hub Models 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
This option is ideal if you need to customize the model beyond the default configuration provided here.
See our repository for MobileNet-v2 on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.image_classification
Model Stats:
- Model checkpoint: Imagenet
- Input resolution: 224x224
- Number of parameters: 3.49M
- Model size (float): 13.3 MB
- Model size (w8a16): 4.39 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| MobileNet-v2 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.277 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | ONNX | float | Snapdragon® 8 Elite Mobile | 0.335 ms | 0 - 34 MB | NPU |
| MobileNet-v2 | ONNX | float | Snapdragon® X2 Elite | 0.305 ms | 7 - 7 MB | NPU |
| MobileNet-v2 | ONNX | float | Snapdragon® X Elite | 0.775 ms | 7 - 7 MB | NPU |
| MobileNet-v2 | ONNX | float | Snapdragon® X Elite | 0.775 ms | 7 - 7 MB | NPU |
| MobileNet-v2 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.422 ms | 0 - 52 MB | NPU |
| MobileNet-v2 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 0.623 ms | 0 - 2 MB | NPU |
| MobileNet-v2 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.335 ms | 0 - 34 MB | NPU |
| MobileNet-v2 | ONNX | float | Qualcomm® QCS9075 | 0.894 ms | 1 - 3 MB | NPU |
| MobileNet-v2 | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.219 ms | 0 - 37 MB | NPU |
| MobileNet-v2 | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 0.294 ms | 0 - 37 MB | NPU |
| MobileNet-v2 | ONNX | w8a16 | Snapdragon® X2 Elite | 0.255 ms | 5 - 5 MB | NPU |
| MobileNet-v2 | ONNX | w8a16 | Snapdragon® X Elite | 0.716 ms | 3 - 3 MB | NPU |
| MobileNet-v2 | ONNX | w8a16 | Snapdragon® X Elite | 0.716 ms | 3 - 3 MB | NPU |
| MobileNet-v2 | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.393 ms | 0 - 47 MB | NPU |
| MobileNet-v2 | ONNX | w8a16 | Qualcomm® QCS6490 | 45.717 ms | 17 - 20 MB | CPU |
| MobileNet-v2 | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 0.577 ms | 0 - 2 MB | NPU |
| MobileNet-v2 | ONNX | w8a16 | Qualcomm® QCM6690 | 18.355 ms | 14 - 21 MB | CPU |
| MobileNet-v2 | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 13.085 ms | 14 - 22 MB | CPU |
| MobileNet-v2 | ONNX | w8a16 | Qualcomm® QCS9075 | 0.778 ms | 0 - 3 MB | NPU |
| MobileNet-v2 | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.294 ms | 0 - 37 MB | NPU |
| MobileNet-v2 | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 13.085 ms | 14 - 22 MB | CPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.278 ms | 0 - 37 MB | NPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Elite Mobile | 0.34 ms | 0 - 38 MB | NPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Snapdragon® X2 Elite | 0.327 ms | 6 - 6 MB | NPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Snapdragon® X Elite | 0.782 ms | 4 - 4 MB | NPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Snapdragon® X Elite | 0.782 ms | 4 - 4 MB | NPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 0.46 ms | 0 - 47 MB | NPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Qualcomm® QCS6490 | 46.8 ms | 14 - 17 MB | CPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Qualcomm® QCS8550 (Proxy) | 0.637 ms | 0 - 2 MB | NPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.34 ms | 0 - 38 MB | NPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Qualcomm® QCM6690 | 18.456 ms | 12 - 19 MB | CPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 12.989 ms | 14 - 22 MB | CPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Qualcomm® QCS9075 | 0.809 ms | 0 - 3 MB | NPU |
| MobileNet-v2 | ONNX | w8a16_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 12.989 ms | 14 - 22 MB | CPU |
| MobileNet-v2 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.254 ms | 0 - 36 MB | NPU |
| MobileNet-v2 | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 0.303 ms | 0 - 30 MB | NPU |
| MobileNet-v2 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.273 ms | 5 - 5 MB | NPU |
| MobileNet-v2 | ONNX | w8a8 | Snapdragon® X Elite | 0.627 ms | 3 - 3 MB | NPU |
| MobileNet-v2 | ONNX | w8a8 | Snapdragon® X Elite | 0.627 ms | 3 - 3 MB | NPU |
| MobileNet-v2 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.354 ms | 0 - 45 MB | NPU |
| MobileNet-v2 | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.483 ms | 0 - 2 MB | NPU |
| MobileNet-v2 | ONNX | w8a8 | Qualcomm® QCS9075 | 0.64 ms | 0 - 3 MB | NPU |
| MobileNet-v2 | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.303 ms | 0 - 30 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.343 ms | 1 - 34 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 0.454 ms | 0 - 30 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Snapdragon® X2 Elite | 0.49 ms | 1 - 1 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Snapdragon® X Elite | 1.078 ms | 1 - 1 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Snapdragon® X Elite | 1.078 ms | 1 - 1 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.601 ms | 0 - 50 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 0.92 ms | 1 - 2 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Qualcomm® SA8775P | 1.207 ms | 1 - 32 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Qualcomm® SA8775P | 1.207 ms | 1 - 32 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Qualcomm® SA8775P | 1.207 ms | 1 - 32 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 1.702 ms | 0 - 52 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Qualcomm® SA7255P | 2.671 ms | 1 - 29 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.454 ms | 0 - 30 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Qualcomm® SA8295P | 1.498 ms | 1 - 29 MB | NPU |
| MobileNet-v2 | QNN_DLC | float | Qualcomm® QCS9075 | 1.124 ms | 1 - 3 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.297 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 0.373 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 0.435 ms | 0 - 0 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Snapdragon® X Elite | 0.962 ms | 0 - 0 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Snapdragon® X Elite | 0.962 ms | 0 - 0 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.561 ms | 0 - 41 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Qualcomm® QCS6490 | 2.304 ms | 0 - 2 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 0.807 ms | 0 - 6 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.001 ms | 0 - 32 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.001 ms | 0 - 32 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.001 ms | 0 - 32 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 3.343 ms | 0 - 143 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Qualcomm® QCS8450 (Proxy) | 0.981 ms | 0 - 44 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 0.864 ms | 0 - 141 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 0.984 ms | 0 - 2 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Qualcomm® SA7255P | 1.778 ms | 0 - 29 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.373 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Qualcomm® SA8295P | 1.275 ms | 0 - 28 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 0.864 ms | 0 - 141 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.322 ms | 0 - 34 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Elite Mobile | 0.415 ms | 0 - 34 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® X2 Elite | 0.466 ms | 0 - 0 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® X Elite | 1.018 ms | 0 - 0 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® X Elite | 1.018 ms | 0 - 0 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 0.599 ms | 0 - 43 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® QCS8550 (Proxy) | 0.85 ms | 0 - 2 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA8775P | 1.05 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA8775P | 1.05 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA8775P | 1.05 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 8 Elite For Galaxy Mobile | 0.415 ms | 0 - 34 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® QCM6690 | 4.752 ms | 0 - 144 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® SA7255P | 2.022 ms | 0 - 30 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 0.984 ms | 0 - 142 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Qualcomm® QCS9075 | 1.021 ms | 0 - 2 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a16_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 0.984 ms | 0 - 142 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.177 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 0.228 ms | 0 - 29 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.279 ms | 0 - 0 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.58 ms | 0 - 0 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.58 ms | 0 - 0 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.317 ms | 0 - 41 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 1.393 ms | 2 - 4 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.453 ms | 0 - 23 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.627 ms | 0 - 31 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.627 ms | 0 - 31 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.627 ms | 0 - 31 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.544 ms | 0 - 2 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.626 ms | 0 - 43 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.476 ms | 0 - 28 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 1.78 ms | 0 - 29 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.228 ms | 0 - 29 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 1.053 ms | 0 - 28 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 0.808 ms | 0 - 27 MB | NPU |
| MobileNet-v2 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.476 ms | 0 - 28 MB | NPU |
| MobileNet-v2 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.344 ms | 0 - 37 MB | NPU |
| MobileNet-v2 | TFLITE | float | Snapdragon® 8 Elite Mobile | 0.45 ms | 0 - 37 MB | NPU |
| MobileNet-v2 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.6 ms | 0 - 53 MB | NPU |
| MobileNet-v2 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 0.925 ms | 0 - 2 MB | NPU |
| MobileNet-v2 | TFLITE | float | Qualcomm® SA8775P | 1.225 ms | 0 - 35 MB | NPU |
| MobileNet-v2 | TFLITE | float | Qualcomm® SA8775P | 1.225 ms | 0 - 35 MB | NPU |
| MobileNet-v2 | TFLITE | float | Qualcomm® SA8775P | 1.225 ms | 0 - 35 MB | NPU |
| MobileNet-v2 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 1.708 ms | 0 - 54 MB | NPU |
| MobileNet-v2 | TFLITE | float | Qualcomm® SA7255P | 2.706 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 0.45 ms | 0 - 37 MB | NPU |
| MobileNet-v2 | TFLITE | float | Qualcomm® SA8295P | 1.499 ms | 0 - 31 MB | NPU |
| MobileNet-v2 | TFLITE | float | Qualcomm® QCS9075 | 1.129 ms | 0 - 10 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.181 ms | 0 - 36 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 0.216 ms | 0 - 30 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.291 ms | 0 - 44 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Qualcomm® QCS6490 | 1.098 ms | 0 - 7 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.407 ms | 0 - 2 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Qualcomm® SA8775P | 0.602 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Qualcomm® SA8775P | 0.602 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Qualcomm® SA8775P | 0.602 ms | 0 - 33 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.574 ms | 0 - 6 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 0.549 ms | 0 - 46 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.412 ms | 0 - 30 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Qualcomm® QCM6690 | 1.651 ms | 0 - 29 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.216 ms | 0 - 30 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Qualcomm® SA7255P | 0.97 ms | 0 - 32 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Qualcomm® SA8295P | 0.802 ms | 0 - 29 MB | NPU |
| MobileNet-v2 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.412 ms | 0 - 30 MB | NPU |
License
- The license for the original implementation of MobileNet-v2 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.
