Hugging Face's logo

qualcomm
/
MobileNet-v3-Large

Image Classification
PyTorch
backbone
bu_auto
real_time
android
Model card Files
xet
 Community

MobileNet-v3-Large: Optimized for Qualcomm Devices

MobileNet-v3-Large 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-v3-Large 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
QNN_DLC float Universal QAIRT 2.45 Download
QNN_DLC w8a16 Universal QAIRT 2.45 Download
TFLITE float Universal QAIRT 2.45 Download

For more device-specific assets and performance metrics, visit MobileNet-v3-Large 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-v3-Large 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: 5.47M
  • Model size (float): 20.9 MB
  • Model size (w8a16): 6.35 MB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
MobileNet-v3-Large ONNX float Snapdragon® X2 Elite 0.33 ms 2 - 2 MB NPU
MobileNet-v3-Large ONNX float Snapdragon® X Elite 0.72 ms 12 - 12 MB NPU
MobileNet-v3-Large ONNX float Snapdragon® 8 Gen 3 Mobile 0.491 ms 0 - 57 MB NPU
MobileNet-v3-Large ONNX float Snapdragon® 8 Gen 1 Mobile 1.212 ms 1 - 59 MB NPU
MobileNet-v3-Large ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.713 ms 0 - 15 MB NPU
MobileNet-v3-Large ONNX float Qualcomm® QCS8450 1.212 ms 1 - 59 MB NPU
MobileNet-v3-Large ONNX float Qualcomm® Dragonwing™ IQ-9075 0.961 ms 1 - 3 MB NPU
MobileNet-v3-Large ONNX float Qualcomm® Dragonwing™ IQ-X7181 0.72 ms 12 - 12 MB NPU
MobileNet-v3-Large ONNX float Qualcomm® Dragonwing™ Q-8750 0.4 ms 0 - 30 MB NPU
MobileNet-v3-Large ONNX float Snapdragon® 8 Elite Mobile 0.4 ms 0 - 30 MB NPU
MobileNet-v3-Large ONNX float Snapdragon® 8 Elite Gen 5 Mobile 0.36 ms 0 - 31 MB NPU
MobileNet-v3-Large ONNX w8a16 Snapdragon® X2 Elite 0.31 ms 1 - 1 MB NPU
MobileNet-v3-Large ONNX w8a16 Snapdragon® X Elite 0.714 ms 7 - 7 MB NPU
MobileNet-v3-Large ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 0.479 ms 0 - 67 MB NPU
MobileNet-v3-Large ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 0.972 ms 0 - 69 MB NPU
MobileNet-v3-Large ONNX w8a16 Qualcomm® Dragonwing™ QCS6490 2.857 ms 0 - 3 MB NPU
MobileNet-v3-Large ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.698 ms 0 - 9 MB NPU
MobileNet-v3-Large ONNX w8a16 Qualcomm® QCS8450 0.972 ms 0 - 69 MB NPU
MobileNet-v3-Large ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 0.836 ms 0 - 3 MB NPU
MobileNet-v3-Large ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 0.714 ms 7 - 7 MB NPU
MobileNet-v3-Large ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 0.356 ms 0 - 42 MB NPU
MobileNet-v3-Large ONNX w8a16 Snapdragon® 8 Elite Mobile 0.356 ms 0 - 42 MB NPU
MobileNet-v3-Large ONNX w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.29 ms 0 - 48 MB NPU
MobileNet-v3-Large QNN_DLC float Snapdragon® X2 Elite 0.517 ms 1 - 1 MB NPU
MobileNet-v3-Large QNN_DLC float Snapdragon® X Elite 1.174 ms 1 - 1 MB NPU
MobileNet-v3-Large QNN_DLC float Snapdragon® 8 Gen 3 Mobile 0.685 ms 0 - 51 MB NPU
MobileNet-v3-Large QNN_DLC float Snapdragon® 8 Gen 1 Mobile 1.889 ms 0 - 58 MB NPU
MobileNet-v3-Large QNN_DLC float Qualcomm® Dragonwing™ QCS8275 2.918 ms 0 - 30 MB NPU
MobileNet-v3-Large QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.027 ms 1 - 2 MB NPU
MobileNet-v3-Large QNN_DLC float Qualcomm® SA8775P 1.368 ms 1 - 33 MB NPU
MobileNet-v3-Large QNN_DLC float Qualcomm® SA8650P 1.368 ms 1 - 33 MB NPU
MobileNet-v3-Large QNN_DLC float Qualcomm® SA8255P 1.368 ms 1 - 33 MB NPU
MobileNet-v3-Large QNN_DLC float Qualcomm® QCS8450 1.889 ms 0 - 58 MB NPU
MobileNet-v3-Large QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 1.223 ms 3 - 5 MB NPU
MobileNet-v3-Large QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 1.174 ms 1 - 1 MB NPU
MobileNet-v3-Large QNN_DLC float Qualcomm® Dragonwing™ Q-8750 0.527 ms 0 - 35 MB NPU
MobileNet-v3-Large QNN_DLC float Qualcomm® SA7255P 2.918 ms 0 - 30 MB NPU
MobileNet-v3-Large QNN_DLC float Qualcomm® SA8295P 1.847 ms 0 - 31 MB NPU
MobileNet-v3-Large QNN_DLC float Snapdragon® 8 Elite Mobile 0.527 ms 0 - 35 MB NPU
MobileNet-v3-Large QNN_DLC float Snapdragon® 8 Elite Gen 5 Mobile 0.428 ms 1 - 32 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Snapdragon® X2 Elite 0.508 ms 0 - 0 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Snapdragon® X Elite 1.092 ms 0 - 0 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 0.659 ms 0 - 52 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Snapdragon® 8 Gen 1 Mobile 1.223 ms 0 - 59 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS6490 3.369 ms 0 - 2 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8275 2.071 ms 0 - 36 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.95 ms 0 - 2 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® SA8775P 1.172 ms 0 - 39 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® SA8650P 1.172 ms 0 - 39 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® SA8255P 1.172 ms 0 - 39 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® QCS8450 1.223 ms 0 - 59 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 1.034 ms 2 - 4 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 1.092 ms 0 - 0 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-6690 4.344 ms 0 - 151 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-7790 1.017 ms 0 - 37 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 0.464 ms 0 - 35 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® SA7255P 2.071 ms 0 - 36 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Qualcomm® SA8295P 1.586 ms 0 - 37 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Snapdragon® 8 Elite Mobile 0.464 ms 0 - 35 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 Mobile 0.367 ms 0 - 36 MB NPU
MobileNet-v3-Large QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 1.017 ms 0 - 37 MB NPU
MobileNet-v3-Large TFLITE float Snapdragon® 8 Gen 3 Mobile 0.696 ms 0 - 56 MB NPU
MobileNet-v3-Large TFLITE float Snapdragon® 8 Gen 1 Mobile 1.915 ms 0 - 66 MB NPU
MobileNet-v3-Large TFLITE float Qualcomm® Dragonwing™ QCS8275 3.078 ms 0 - 35 MB NPU
MobileNet-v3-Large TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.039 ms 0 - 2 MB NPU
MobileNet-v3-Large TFLITE float Qualcomm® SA8775P 1.415 ms 0 - 38 MB NPU
MobileNet-v3-Large TFLITE float Qualcomm® SA8650P 1.415 ms 0 - 38 MB NPU
MobileNet-v3-Large TFLITE float Qualcomm® SA8255P 1.415 ms 0 - 38 MB NPU
MobileNet-v3-Large TFLITE float Qualcomm® QCS8450 1.915 ms 0 - 66 MB NPU
MobileNet-v3-Large TFLITE float Qualcomm® Dragonwing™ IQ-9075 1.281 ms 0 - 15 MB NPU
MobileNet-v3-Large TFLITE float Qualcomm® Dragonwing™ Q-8750 0.528 ms 0 - 41 MB NPU
MobileNet-v3-Large TFLITE float Qualcomm® SA7255P 3.078 ms 0 - 35 MB NPU
MobileNet-v3-Large TFLITE float Qualcomm® SA8295P 1.864 ms 0 - 35 MB NPU
MobileNet-v3-Large TFLITE float Snapdragon® 8 Elite Mobile 0.528 ms 0 - 41 MB NPU
MobileNet-v3-Large TFLITE float Snapdragon® 8 Elite Gen 5 Mobile 0.436 ms 0 - 36 MB NPU

License

  • The license for the original implementation of MobileNet-v3-Large can be found here.

References

Community

Downloads last month

-

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

Paper for qualcomm/MobileNet-v3-Large