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 2.45, ONNX Runtime 1.27.1 Download
ONNX w8a16 Universal QAIRT 2.45, ONNX Runtime 1.27.1 Download
ONNX w8a16_mixed_int16 Universal QAIRT 2.45, ONNX Runtime 1.27.1 Download
ONNX w8a8 Universal QAIRT 2.45, ONNX Runtime 1.27.1 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® X2 Elite 0.306 ms 2 - 2 MB NPU
MobileNet-v2 ONNX float Snapdragon® X Elite 0.635 ms 7 - 7 MB NPU
MobileNet-v2 ONNX float Snapdragon® 8 Gen 3 Mobile 0.428 ms 0 - 51 MB NPU
MobileNet-v2 ONNX float Snapdragon® 8 Gen 1 Mobile 0.95 ms 0 - 56 MB NPU
MobileNet-v2 ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.634 ms 0 - 9 MB NPU
MobileNet-v2 ONNX float Qualcomm® QCS8450 0.95 ms 0 - 56 MB NPU
MobileNet-v2 ONNX float Qualcomm® Dragonwing™ IQ-9075 0.853 ms 1 - 3 MB NPU
MobileNet-v2 ONNX float Qualcomm® Dragonwing™ IQ-X7181 0.635 ms 7 - 7 MB NPU
MobileNet-v2 ONNX float Qualcomm® Dragonwing™ Q-8750 0.336 ms 0 - 28 MB NPU
MobileNet-v2 ONNX float Snapdragon® 8 Elite Mobile 0.336 ms 0 - 28 MB NPU
MobileNet-v2 ONNX float Snapdragon® 8 Elite Gen 5 Mobile 0.279 ms 0 - 29 MB NPU
MobileNet-v2 ONNX w8a16 Snapdragon® X2 Elite 0.241 ms 1 - 1 MB NPU
MobileNet-v2 ONNX w8a16 Snapdragon® X Elite 0.558 ms 0 - 0 MB NPU
MobileNet-v2 ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 0.38 ms 0 - 51 MB NPU
MobileNet-v2 ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 0.709 ms 0 - 56 MB NPU
MobileNet-v2 ONNX w8a16 Qualcomm® Dragonwing™ QCS6490 2.187 ms 0 - 3 MB NPU
MobileNet-v2 ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.562 ms 0 - 87 MB NPU
MobileNet-v2 ONNX w8a16 Qualcomm® QCS8450 0.709 ms 0 - 56 MB NPU
MobileNet-v2 ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 0.708 ms 0 - 3 MB NPU
MobileNet-v2 ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 0.558 ms 0 - 0 MB NPU
MobileNet-v2 ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 0.277 ms 0 - 35 MB NPU
MobileNet-v2 ONNX w8a16 Snapdragon® 8 Elite Mobile 0.277 ms 0 - 35 MB NPU
MobileNet-v2 ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.217 ms 0 - 36 MB NPU
MobileNet-v2 ONNX w8a16_mixed_int16 Snapdragon® X2 Elite 0.29 ms 1 - 1 MB NPU
MobileNet-v2 ONNX w8a16_mixed_int16 Snapdragon® X Elite 0.622 ms 0 - 0 MB NPU
MobileNet-v2 ONNX w8a16_mixed_int16 Snapdragon® 8 Gen 3 Mobile 0.426 ms 0 - 52 MB NPU
MobileNet-v2 ONNX w8a16_mixed_int16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.612 ms 0 - 7 MB NPU
MobileNet-v2 ONNX w8a16_mixed_int16 Qualcomm® Dragonwing™ IQ-9075 0.736 ms 0 - 3 MB NPU
MobileNet-v2 ONNX w8a16_mixed_int16 Qualcomm® Dragonwing™ IQ-X7181 0.622 ms 0 - 0 MB NPU
MobileNet-v2 ONNX w8a16_mixed_int16 Qualcomm® Dragonwing™ Q-8750 0.327 ms 0 - 37 MB NPU
MobileNet-v2 ONNX w8a16_mixed_int16 Snapdragon® 8 Elite Mobile 0.327 ms 0 - 37 MB NPU
MobileNet-v2 ONNX w8a16_mixed_int16 Snapdragon® 8 Elite Gen 5 Mobile 0.265 ms 0 - 37 MB NPU
MobileNet-v2 ONNX w8a8 Snapdragon® X2 Elite 0.254 ms 1 - 1 MB NPU
MobileNet-v2 ONNX w8a8 Snapdragon® X Elite 0.459 ms 0 - 0 MB NPU
MobileNet-v2 ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 0.328 ms 0 - 50 MB NPU
MobileNet-v2 ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 0.587 ms 0 - 55 MB NPU
MobileNet-v2 ONNX w8a8 Qualcomm® Dragonwing™ QCS6490 1.293 ms 0 - 3 MB NPU
MobileNet-v2 ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.46 ms 0 - 31 MB NPU
MobileNet-v2 ONNX w8a8 Qualcomm® QCS8450 0.587 ms 0 - 55 MB NPU
MobileNet-v2 ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 0.575 ms 0 - 3 MB NPU
MobileNet-v2 ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 0.459 ms 0 - 0 MB NPU
MobileNet-v2 ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 0.286 ms 0 - 39 MB NPU
MobileNet-v2 ONNX w8a8 Snapdragon® 8 Elite Mobile 0.286 ms 0 - 39 MB NPU
MobileNet-v2 ONNX w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.252 ms 0 - 34 MB NPU
MobileNet-v2 QNN_DLC float Snapdragon® X2 Elite 0.659 ms 1 - 1 MB NPU
MobileNet-v2 QNN_DLC float Snapdragon® X Elite 1.084 ms 1 - 1 MB NPU
MobileNet-v2 QNN_DLC float Snapdragon® 8 Gen 3 Mobile 0.603 ms 0 - 49 MB NPU
MobileNet-v2 QNN_DLC float Snapdragon® 8 Gen 1 Mobile 1.693 ms 0 - 55 MB NPU
MobileNet-v2 QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.924 ms 1 - 2 MB NPU
MobileNet-v2 QNN_DLC float Qualcomm® SA8775P 1.212 ms 1 - 32 MB NPU
MobileNet-v2 QNN_DLC float Qualcomm® SA8650P 1.212 ms 1 - 32 MB NPU
MobileNet-v2 QNN_DLC float Qualcomm® SA8255P 1.212 ms 1 - 32 MB NPU
MobileNet-v2 QNN_DLC float Qualcomm® QCS8450 1.693 ms 0 - 55 MB NPU
MobileNet-v2 QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 1.159 ms 1 - 3 MB NPU
MobileNet-v2 QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 1.084 ms 1 - 1 MB NPU
MobileNet-v2 QNN_DLC float Qualcomm® Dragonwing™ Q-8750 0.449 ms 1 - 35 MB NPU
MobileNet-v2 QNN_DLC float Qualcomm® SA8295P 1.487 ms 0 - 28 MB NPU
MobileNet-v2 QNN_DLC float Snapdragon® 8 Elite Mobile 0.449 ms 1 - 35 MB NPU
MobileNet-v2 QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 0.348 ms 1 - 32 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 0.937 ms 0 - 0 MB NPU
MobileNet-v2 QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 0.566 ms 0 - 43 MB NPU
MobileNet-v2 QNN_DLC w8a16 Snapdragon® 8 Gen 1 Mobile 0.984 ms 0 - 50 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS6490 2.887 ms 2 - 4 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8275 1.792 ms 0 - 31 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.807 ms 0 - 2 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® SA8775P 1.001 ms 0 - 34 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® SA8650P 1.001 ms 0 - 34 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® SA8255P 1.001 ms 0 - 34 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® QCS8450 0.984 ms 0 - 50 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 0.888 ms 2 - 4 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 0.937 ms 0 - 0 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-6690 3.379 ms 0 - 143 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-7790 0.875 ms 0 - 143 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 0.375 ms 0 - 30 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® SA7255P 1.792 ms 0 - 31 MB NPU
MobileNet-v2 QNN_DLC w8a16 Qualcomm® SA8295P 1.287 ms 0 - 30 MB NPU
MobileNet-v2 QNN_DLC w8a16 Snapdragon® 8 Elite Mobile 0.375 ms 0 - 30 MB NPU
MobileNet-v2 QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.303 ms 0 - 31 MB NPU
MobileNet-v2 QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 0.875 ms 0 - 143 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Snapdragon® X2 Elite 0.475 ms 0 - 0 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Snapdragon® X Elite 0.997 ms 0 - 0 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Snapdragon® 8 Gen 3 Mobile 0.607 ms 0 - 44 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.858 ms 0 - 2 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Qualcomm® SA8775P 1.044 ms 0 - 35 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Qualcomm® SA8650P 1.044 ms 0 - 35 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Qualcomm® SA8255P 1.044 ms 0 - 35 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Qualcomm® Dragonwing™ IQ-9075 0.939 ms 2 - 4 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Qualcomm® Dragonwing™ IQ-X7181 0.997 ms 0 - 0 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Qualcomm® Dragonwing™ Q-6690 4.769 ms 0 - 145 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Qualcomm® Dragonwing™ Q-7790 0.981 ms 0 - 144 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Qualcomm® Dragonwing™ Q-8750 0.42 ms 0 - 31 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Snapdragon® 8 Elite Mobile 0.42 ms 0 - 31 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Snapdragon® 8 Elite Gen 5 Mobile 0.327 ms 0 - 33 MB NPU
MobileNet-v2 QNN_DLC w8a16_mixed_int16 Snapdragon® 7 Gen 4 Mobile 0.981 ms 0 - 144 MB NPU
MobileNet-v2 QNN_DLC w8a8 Snapdragon® X2 Elite 0.282 ms 0 - 0 MB NPU
MobileNet-v2 QNN_DLC w8a8 Snapdragon® X Elite 0.554 ms 0 - 0 MB NPU
MobileNet-v2 QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 0.312 ms 0 - 43 MB NPU
MobileNet-v2 QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 0.63 ms 0 - 48 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS6490 1.435 ms 2 - 4 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8275 1.054 ms 0 - 31 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.454 ms 0 - 61 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® SA8775P 0.627 ms 0 - 33 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® SA8650P 0.627 ms 0 - 33 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® SA8255P 0.627 ms 0 - 33 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® QCS8450 0.63 ms 0 - 48 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 0.525 ms 0 - 2 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 0.554 ms 0 - 0 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-6690 1.797 ms 0 - 30 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-7790 0.479 ms 0 - 30 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 0.232 ms 0 - 30 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® SA7255P 1.054 ms 0 - 31 MB NPU
MobileNet-v2 QNN_DLC w8a8 Qualcomm® SA8295P 0.802 ms 0 - 29 MB NPU
MobileNet-v2 QNN_DLC w8a8 Snapdragon® 8 Elite Mobile 0.232 ms 0 - 30 MB NPU
MobileNet-v2 QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.191 ms 0 - 31 MB NPU
MobileNet-v2 QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 0.479 ms 0 - 30 MB NPU
MobileNet-v2 TFLITE float Snapdragon® 8 Gen 3 Mobile 0.603 ms 0 - 52 MB NPU
MobileNet-v2 TFLITE float Snapdragon® 8 Gen 1 Mobile 1.706 ms 0 - 58 MB NPU
MobileNet-v2 TFLITE float Qualcomm® Dragonwing™ QCS8275 2.661 ms 0 - 33 MB NPU
MobileNet-v2 TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.928 ms 0 - 2 MB NPU
MobileNet-v2 TFLITE float Qualcomm® SA8775P 1.23 ms 0 - 35 MB NPU
MobileNet-v2 TFLITE float Qualcomm® SA8650P 1.23 ms 0 - 35 MB NPU
MobileNet-v2 TFLITE float Qualcomm® SA8255P 1.23 ms 0 - 35 MB NPU
MobileNet-v2 TFLITE float Qualcomm® QCS8450 1.706 ms 0 - 58 MB NPU
MobileNet-v2 TFLITE float Qualcomm® Dragonwing™ IQ-9075 1.1 ms 0 - 10 MB NPU
MobileNet-v2 TFLITE float Qualcomm® Dragonwing™ Q-8750 0.449 ms 0 - 32 MB NPU
MobileNet-v2 TFLITE float Qualcomm® SA7255P 2.661 ms 0 - 33 MB NPU
MobileNet-v2 TFLITE float Qualcomm® SA8295P 1.504 ms 0 - 30 MB NPU
MobileNet-v2 TFLITE float Snapdragon® 8 Elite Mobile 0.449 ms 0 - 32 MB NPU
MobileNet-v2 TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 0.347 ms 0 - 34 MB NPU
MobileNet-v2 TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 0.291 ms 0 - 44 MB NPU
MobileNet-v2 TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 0.542 ms 0 - 51 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® Dragonwing™ QCS6490 1.305 ms 0 - 7 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® Dragonwing™ QCS8275 0.975 ms 0 - 33 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.403 ms 0 - 2 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® SA8775P 0.602 ms 0 - 34 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® SA8650P 0.602 ms 0 - 34 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® SA8255P 0.602 ms 0 - 34 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® QCS8450 0.542 ms 0 - 51 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® Dragonwing™ IQ-9075 0.624 ms 0 - 6 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® Dragonwing™ Q-6690 1.645 ms 0 - 29 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® Dragonwing™ Q-7790 0.414 ms 0 - 30 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® Dragonwing™ Q-8750 0.216 ms 0 - 30 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® SA7255P 0.975 ms 0 - 33 MB NPU
MobileNet-v2 TFLITE w8a8 Qualcomm® SA8295P 0.817 ms 0 - 29 MB NPU
MobileNet-v2 TFLITE w8a8 Snapdragon® 8 Elite Mobile 0.216 ms 0 - 30 MB NPU
MobileNet-v2 TFLITE w8a8 Snapdragon® 8 Elite Gen 5 Mobile 0.184 ms 0 - 33 MB NPU
MobileNet-v2 TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 0.414 ms 0 - 30 MB NPU

License

  • The license for the original implementation of MobileNet-v2 can be found here.

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for qualcomm/MobileNet-v2

Finetunes
2 models

Paper for qualcomm/MobileNet-v2