Hugging Face's logo

qualcomm
/
GPUNet

Image Classification
PyTorch
backbone
bu_auto
android
Model card Files
xet
 Community

GPUNet: Optimized for Qualcomm Devices

GPUNet 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 GPUNet 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 2.50, ONNX Runtime 1.27.1 Download
ONNX w8a16 Universal QAIRT 2.50, ONNX Runtime 1.27.1 Download
ONNX w8a8 Universal QAIRT 2.50, ONNX Runtime 1.27.1 Download
QNN_DLC float Universal QAIRT 2.50 Download
QNN_DLC w8a16 Universal QAIRT 2.50 Download
QNN_DLC w8a8 Universal QAIRT 2.50 Download
TFLITE float Universal QAIRT 2.50 Download
TFLITE w8a8 Universal QAIRT 2.50 Download

For more device-specific assets and performance metrics, visit GPUNet 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 GPUNet on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.image_classification

Model Stats:

  • Input resolution: 224x224
  • Model checkpoint: Imagenet
  • Model size (float): 45.28MB
  • Model size (w8a8): 21.3MB
  • Number of parameters: 10.49M

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
GPUNet ONNX float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.513 ms 0 - 38 MB NPU
GPUNet ONNX float Snapdragon® 8 Elite For Galaxy Mobile 0.595 ms 0 - 38 MB NPU
GPUNet ONNX float Snapdragon® X2 Elite 0.479 ms 2 - 2 MB NPU
GPUNet ONNX float Snapdragon® X Elite 0.984 ms 25 - 25 MB NPU
GPUNet ONNX float Snapdragon® 8 Gen 3 Mobile 0.748 ms 0 - 60 MB NPU
GPUNet ONNX float Snapdragon® 8 Gen 1 Mobile 1.64 ms 1 - 68 MB NPU
GPUNet ONNX float Qualcomm® Dragonwing™ IQ-8275 1.501 ms 1 - 5 MB NPU
GPUNet ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.009 ms 0 - 32 MB NPU
GPUNet ONNX float Qualcomm® QCS8450 1.64 ms 1 - 68 MB NPU
GPUNet ONNX float Qualcomm® Dragonwing™ IQ-9075 1.388 ms 1 - 4 MB NPU
GPUNet ONNX float Qualcomm® Dragonwing™ IQ-X7181 0.984 ms 25 - 25 MB NPU
GPUNet ONNX float Qualcomm® Dragonwing™ Q-8750 0.595 ms 0 - 38 MB NPU
GPUNet ONNX w8a16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.372 ms 0 - 52 MB NPU
GPUNet ONNX w8a16 Snapdragon® 8 Elite For Galaxy Mobile 0.43 ms 0 - 55 MB NPU
GPUNet ONNX w8a16 Snapdragon® X2 Elite 0.388 ms 1 - 1 MB NPU
GPUNet ONNX w8a16 Snapdragon® X Elite 0.841 ms 13 - 13 MB NPU
GPUNet ONNX w8a16 Snapdragon® 8 Gen 3 Mobile 0.595 ms 0 - 70 MB NPU
GPUNet ONNX w8a16 Snapdragon® 8 Gen 1 Mobile 1.03 ms 0 - 74 MB NPU
GPUNet ONNX w8a16 Qualcomm® Dragonwing™ QCS6490 2.728 ms 0 - 3 MB NPU
GPUNet ONNX w8a16 Qualcomm® Dragonwing™ IQ-8275 0.895 ms 0 - 4 MB NPU
GPUNet ONNX w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.825 ms 0 - 16 MB NPU
GPUNet ONNX w8a16 Qualcomm® QCS8450 1.03 ms 0 - 74 MB NPU
GPUNet ONNX w8a16 Qualcomm® Dragonwing™ IQ-9075 0.96 ms 0 - 3 MB NPU
GPUNet ONNX w8a16 Qualcomm® Dragonwing™ IQ-X7181 0.841 ms 13 - 13 MB NPU
GPUNet ONNX w8a16 Qualcomm® Dragonwing™ Q-6690 5.745 ms 0 - 168 MB NPU
GPUNet ONNX w8a16 Qualcomm® Dragonwing™ Q-7790 1.084 ms 0 - 52 MB NPU
GPUNet ONNX w8a16 Qualcomm® Dragonwing™ Q-8750 0.43 ms 0 - 55 MB NPU
GPUNet ONNX w8a16 Snapdragon® 7 Gen 4 Mobile 1.084 ms 0 - 52 MB NPU
GPUNet ONNX w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.361 ms 0 - 54 MB NPU
GPUNet ONNX w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.391 ms 0 - 52 MB NPU
GPUNet ONNX w8a8 Snapdragon® X2 Elite 0.3 ms 1 - 1 MB NPU
GPUNet ONNX w8a8 Snapdragon® X Elite 0.584 ms 12 - 12 MB NPU
GPUNet ONNX w8a8 Snapdragon® 8 Gen 3 Mobile 0.458 ms 0 - 66 MB NPU
GPUNet ONNX w8a8 Snapdragon® 8 Gen 1 Mobile 0.759 ms 0 - 69 MB NPU
GPUNet ONNX w8a8 Qualcomm® Dragonwing™ QCS6490 1.821 ms 0 - 3 MB NPU
GPUNet ONNX w8a8 Qualcomm® Dragonwing™ IQ-8275 0.709 ms 0 - 4 MB NPU
GPUNet ONNX w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.612 ms 0 - 17 MB NPU
GPUNet ONNX w8a8 Qualcomm® QCS8450 0.759 ms 0 - 69 MB NPU
GPUNet ONNX w8a8 Qualcomm® Dragonwing™ IQ-9075 0.731 ms 0 - 3 MB NPU
GPUNet ONNX w8a8 Qualcomm® Dragonwing™ IQ-X7181 0.584 ms 12 - 12 MB NPU
GPUNet ONNX w8a8 Qualcomm® Dragonwing™ Q-6690 3.243 ms 0 - 50 MB NPU
GPUNet ONNX w8a8 Qualcomm® Dragonwing™ Q-7790 0.709 ms 0 - 49 MB NPU
GPUNet ONNX w8a8 Qualcomm® Dragonwing™ Q-8750 0.391 ms 0 - 52 MB NPU
GPUNet ONNX w8a8 Snapdragon® 7 Gen 4 Mobile 0.709 ms 0 - 49 MB NPU
GPUNet QNN_DLC float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.567 ms 1 - 36 MB NPU
GPUNet QNN_DLC float Snapdragon® 8 Elite For Galaxy Mobile 0.7 ms 0 - 34 MB NPU
GPUNet QNN_DLC float Snapdragon® X2 Elite 0.629 ms 1 - 1 MB NPU
GPUNet QNN_DLC float Snapdragon® X Elite 1.389 ms 1 - 1 MB NPU
GPUNet QNN_DLC float Snapdragon® 8 Gen 3 Mobile 0.903 ms 0 - 59 MB NPU
GPUNet QNN_DLC float Snapdragon® 8 Gen 1 Mobile 2.244 ms 1 - 68 MB NPU
GPUNet QNN_DLC float Qualcomm® Dragonwing™ IQ-8275 1.64 ms 1 - 4 MB NPU
GPUNet QNN_DLC float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.284 ms 1 - 2 MB NPU
GPUNet QNN_DLC float Qualcomm® SA8775P 1.696 ms 1 - 39 MB NPU
GPUNet QNN_DLC float Qualcomm® SA8650P 1.696 ms 1 - 39 MB NPU
GPUNet QNN_DLC float Qualcomm® SA8255P 1.696 ms 1 - 39 MB NPU
GPUNet QNN_DLC float Qualcomm® QCS8450 2.244 ms 1 - 68 MB NPU
GPUNet QNN_DLC float Qualcomm® Dragonwing™ IQ-9075 1.577 ms 1 - 3 MB NPU
GPUNet QNN_DLC float Qualcomm® Dragonwing™ IQ-X7181 1.389 ms 1 - 1 MB NPU
GPUNet QNN_DLC float Qualcomm® Dragonwing™ Q-8750 0.7 ms 0 - 34 MB NPU
GPUNet QNN_DLC float Qualcomm® SA8295P 2.151 ms 0 - 35 MB NPU
GPUNet QNN_DLC w8a16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.446 ms 0 - 47 MB NPU
GPUNet QNN_DLC w8a16 Snapdragon® 8 Elite For Galaxy Mobile 0.537 ms 0 - 47 MB NPU
GPUNet QNN_DLC w8a16 Snapdragon® X2 Elite 0.56 ms 0 - 0 MB NPU
GPUNet QNN_DLC w8a16 Snapdragon® X Elite 1.238 ms 0 - 0 MB NPU
GPUNet QNN_DLC w8a16 Snapdragon® 8 Gen 3 Mobile 0.773 ms 21 - 86 MB NPU
GPUNet QNN_DLC w8a16 Snapdragon® 8 Gen 1 Mobile 1.29 ms 0 - 72 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS6490 3.352 ms 2 - 5 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-8275 1.088 ms 0 - 3 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.071 ms 0 - 2 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® SA8650P 1.282 ms 0 - 45 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® SA8255P 1.282 ms 0 - 45 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® QCS8450 1.29 ms 0 - 72 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-9075 1.165 ms 0 - 3 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® Dragonwing™ IQ-X7181 1.238 ms 0 - 0 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-6690 6.484 ms 0 - 164 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-7790 1.29 ms 0 - 48 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® Dragonwing™ Q-8750 0.537 ms 0 - 47 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® SA7255P 2.459 ms 0 - 48 MB NPU
GPUNet QNN_DLC w8a16 Qualcomm® SA8295P 1.564 ms 0 - 45 MB NPU
GPUNet QNN_DLC w8a16 Snapdragon® 7 Gen 4 Mobile 1.29 ms 0 - 48 MB NPU
GPUNet QNN_DLC w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.295 ms 0 - 46 MB NPU
GPUNet QNN_DLC w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.341 ms 0 - 47 MB NPU
GPUNet QNN_DLC w8a8 Snapdragon® X2 Elite 0.4 ms 0 - 0 MB NPU
GPUNet QNN_DLC w8a8 Snapdragon® X Elite 0.71 ms 0 - 0 MB NPU
GPUNet QNN_DLC w8a8 Snapdragon® 8 Gen 3 Mobile 0.459 ms 0 - 62 MB NPU
GPUNet QNN_DLC w8a8 Snapdragon® 8 Gen 1 Mobile 0.753 ms 0 - 68 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS6490 2.029 ms 0 - 2 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-8275 0.69 ms 0 - 3 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.603 ms 0 - 2 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® SA8775P 0.795 ms 0 - 46 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® SA8650P 0.795 ms 0 - 46 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® SA8255P 0.795 ms 0 - 46 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® QCS8450 0.753 ms 0 - 68 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-9075 0.696 ms 0 - 2 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® Dragonwing™ IQ-X7181 0.71 ms 0 - 0 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-6690 3.495 ms 0 - 46 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-7790 0.791 ms 0 - 45 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® Dragonwing™ Q-8750 0.341 ms 0 - 47 MB NPU
GPUNet QNN_DLC w8a8 Qualcomm® SA8295P 0.982 ms 0 - 42 MB NPU
GPUNet QNN_DLC w8a8 Snapdragon® 7 Gen 4 Mobile 0.791 ms 0 - 45 MB NPU
GPUNet TFLITE float Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.56 ms 0 - 48 MB NPU
GPUNet TFLITE float Snapdragon® 8 Elite For Galaxy Mobile 0.702 ms 0 - 51 MB NPU
GPUNet TFLITE float Snapdragon® 8 Gen 3 Mobile 0.909 ms 0 - 78 MB NPU
GPUNet TFLITE float Snapdragon® 8 Gen 1 Mobile 2.231 ms 0 - 88 MB NPU
GPUNet TFLITE float Qualcomm® Dragonwing™ IQ-8275 1.641 ms 0 - 28 MB NPU
GPUNet TFLITE float Qualcomm® Dragonwing™ QCS8550 (Proxy) 1.252 ms 0 - 3 MB NPU
GPUNet TFLITE float Qualcomm® QCS8450 2.231 ms 0 - 88 MB NPU
GPUNet TFLITE float Qualcomm® Dragonwing™ IQ-9075 1.556 ms 0 - 27 MB NPU
GPUNet TFLITE float Qualcomm® Dragonwing™ Q-8750 0.702 ms 0 - 51 MB NPU
GPUNet TFLITE float Qualcomm® SA7255P 4.691 ms 0 - 48 MB NPU
GPUNet TFLITE float Qualcomm® SA8295P 2.156 ms 0 - 47 MB NPU
GPUNet TFLITE w8a8 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 0.235 ms 0 - 44 MB NPU
GPUNet TFLITE w8a8 Snapdragon® 8 Elite For Galaxy Mobile 0.26 ms 0 - 41 MB NPU
GPUNet TFLITE w8a8 Snapdragon® 8 Gen 3 Mobile 0.332 ms 0 - 60 MB NPU
GPUNet TFLITE w8a8 Snapdragon® 8 Gen 1 Mobile 0.557 ms 0 - 68 MB NPU
GPUNet TFLITE w8a8 Qualcomm® Dragonwing™ QCS6490 1.493 ms 0 - 16 MB NPU
GPUNet TFLITE w8a8 Qualcomm® Dragonwing™ IQ-8275 0.529 ms 0 - 15 MB NPU
GPUNet TFLITE w8a8 Qualcomm® Dragonwing™ QCS8550 (Proxy) 0.421 ms 0 - 2 MB NPU
GPUNet TFLITE w8a8 Qualcomm® SA8775P 0.617 ms 0 - 47 MB NPU
GPUNet TFLITE w8a8 Qualcomm® SA8650P 0.617 ms 0 - 47 MB NPU
GPUNet TFLITE w8a8 Qualcomm® SA8255P 0.617 ms 0 - 47 MB NPU
GPUNet TFLITE w8a8 Qualcomm® QCS8450 0.557 ms 0 - 68 MB NPU
GPUNet TFLITE w8a8 Qualcomm® Dragonwing™ IQ-9075 0.515 ms 0 - 14 MB NPU
GPUNet TFLITE w8a8 Qualcomm® Dragonwing™ Q-6690 2.959 ms 0 - 44 MB NPU
GPUNet TFLITE w8a8 Qualcomm® Dragonwing™ Q-7790 0.61 ms 0 - 44 MB NPU
GPUNet TFLITE w8a8 Qualcomm® Dragonwing™ Q-8750 0.26 ms 0 - 41 MB NPU
GPUNet TFLITE w8a8 Qualcomm® SA7255P 1.082 ms 0 - 44 MB NPU
GPUNet TFLITE w8a8 Qualcomm® SA8295P 0.781 ms 0 - 41 MB NPU
GPUNet TFLITE w8a8 Snapdragon® 7 Gen 4 Mobile 0.61 ms 0 - 44 MB NPU

License

  • The license for the original implementation of GPUNet 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/GPUNet