WideResNet50: Optimized for Qualcomm Devices

WideResNet50 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 WideResNet50 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 1.24.1	Download
ONNX	w8a8	Universal	QAIRT 2.42, ONNX Runtime 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
QNN_DLC	w8a8	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.17.0	Download
TFLITE	w8a8	Universal	QAIRT 2.43, TFLite 2.17.0	Download

For more device-specific assets and performance metrics, visit WideResNet50 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 WideResNet50 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: 68.9M
• Model size (float): 263 MB
• Model size (w8a8): 66.6 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
WideResNet50	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	2.644 ms	1 - 189 MB	NPU
WideResNet50	ONNX	float	Snapdragon® X2 Elite	2.216 ms	132 - 132 MB	NPU
WideResNet50	ONNX	float	Snapdragon® X Elite	4.437 ms	132 - 132 MB	NPU
WideResNet50	ONNX	float	Snapdragon® 8 Gen 3 Mobile	3.464 ms	0 - 226 MB	NPU
WideResNet50	ONNX	float	Qualcomm® QCS8550 (Proxy)	4.563 ms	1 - 69 MB	NPU
WideResNet50	ONNX	float	Qualcomm® QCS9075	6.767 ms	0 - 4 MB	NPU
WideResNet50	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	2.806 ms	0 - 179 MB	NPU
WideResNet50	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.155 ms	0 - 53 MB	NPU
WideResNet50	ONNX	w8a8	Snapdragon® X2 Elite	0.831 ms	66 - 66 MB	NPU
WideResNet50	ONNX	w8a8	Snapdragon® X Elite	1.79 ms	66 - 66 MB	NPU
WideResNet50	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	1.402 ms	0 - 131 MB	NPU
WideResNet50	ONNX	w8a8	Qualcomm® QCS6490	76.461 ms	10 - 107 MB	CPU
WideResNet50	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	1.861 ms	0 - 353 MB	NPU
WideResNet50	ONNX	w8a8	Qualcomm® QCS9075	1.916 ms	0 - 3 MB	NPU
WideResNet50	ONNX	w8a8	Qualcomm® QCM6690	62.051 ms	8 - 17 MB	CPU
WideResNet50	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.229 ms	0 - 55 MB	NPU
WideResNet50	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	57.438 ms	8 - 17 MB	CPU
WideResNet50	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	2.542 ms	0 - 171 MB	NPU
WideResNet50	QNN_DLC	float	Snapdragon® X2 Elite	2.48 ms	1 - 1 MB	NPU
WideResNet50	QNN_DLC	float	Snapdragon® X Elite	4.737 ms	1 - 1 MB	NPU
WideResNet50	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	3.658 ms	1 - 229 MB	NPU
WideResNet50	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	24.59 ms	1 - 175 MB	NPU
WideResNet50	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	4.854 ms	1 - 2 MB	NPU
WideResNet50	QNN_DLC	float	Qualcomm® SA8775P	7.185 ms	1 - 170 MB	NPU
WideResNet50	QNN_DLC	float	Qualcomm® QCS9075	7.076 ms	1 - 3 MB	NPU
WideResNet50	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	9.204 ms	0 - 199 MB	NPU
WideResNet50	QNN_DLC	float	Qualcomm® SA7255P	24.59 ms	1 - 175 MB	NPU
WideResNet50	QNN_DLC	float	Qualcomm® SA8295P	7.895 ms	1 - 152 MB	NPU
WideResNet50	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	2.919 ms	0 - 175 MB	NPU
WideResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.096 ms	0 - 54 MB	NPU
WideResNet50	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.884 ms	0 - 0 MB	NPU
WideResNet50	QNN_DLC	w8a8	Snapdragon® X Elite	1.788 ms	0 - 0 MB	NPU
WideResNet50	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	1.414 ms	0 - 125 MB	NPU
WideResNet50	QNN_DLC	w8a8	Qualcomm® QCS6490	7.783 ms	0 - 2 MB	NPU
WideResNet50	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	3.984 ms	0 - 52 MB	NPU
WideResNet50	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	1.907 ms	0 - 12 MB	NPU
WideResNet50	QNN_DLC	w8a8	Qualcomm® SA8775P	2.018 ms	0 - 52 MB	NPU
WideResNet50	QNN_DLC	w8a8	Qualcomm® QCS9075	1.904 ms	0 - 2 MB	NPU
WideResNet50	QNN_DLC	w8a8	Qualcomm® QCM6690	18.611 ms	0 - 212 MB	NPU
WideResNet50	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	2.529 ms	0 - 127 MB	NPU
WideResNet50	QNN_DLC	w8a8	Qualcomm® SA7255P	3.984 ms	0 - 52 MB	NPU
WideResNet50	QNN_DLC	w8a8	Qualcomm® SA8295P	2.754 ms	0 - 49 MB	NPU
WideResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.146 ms	0 - 50 MB	NPU
WideResNet50	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	2.87 ms	0 - 210 MB	NPU
WideResNet50	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	2.501 ms	0 - 118 MB	NPU
WideResNet50	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	3.569 ms	0 - 255 MB	NPU
WideResNet50	TFLITE	float	Qualcomm® QCS8275 (Proxy)	23.65 ms	0 - 115 MB	NPU
WideResNet50	TFLITE	float	Qualcomm® QCS8550 (Proxy)	4.764 ms	0 - 3 MB	NPU
WideResNet50	TFLITE	float	Qualcomm® SA8775P	29.432 ms	0 - 115 MB	NPU
WideResNet50	TFLITE	float	Qualcomm® QCS9075	7.065 ms	0 - 134 MB	NPU
WideResNet50	TFLITE	float	Qualcomm® QCS8450 (Proxy)	8.18 ms	0 - 228 MB	NPU
WideResNet50	TFLITE	float	Qualcomm® SA7255P	23.65 ms	0 - 115 MB	NPU
WideResNet50	TFLITE	float	Qualcomm® SA8295P	7.677 ms	0 - 102 MB	NPU
WideResNet50	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	2.849 ms	0 - 115 MB	NPU
WideResNet50	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	1.05 ms	0 - 51 MB	NPU
WideResNet50	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	1.29 ms	0 - 128 MB	NPU
WideResNet50	TFLITE	w8a8	Qualcomm® QCS6490	6.851 ms	0 - 68 MB	NPU
WideResNet50	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	3.733 ms	0 - 49 MB	NPU
WideResNet50	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	1.72 ms	0 - 249 MB	NPU
WideResNet50	TFLITE	w8a8	Qualcomm® SA8775P	1.898 ms	0 - 51 MB	NPU
WideResNet50	TFLITE	w8a8	Qualcomm® QCS9075	1.74 ms	0 - 68 MB	NPU
WideResNet50	TFLITE	w8a8	Qualcomm® QCM6690	17.07 ms	0 - 209 MB	NPU
WideResNet50	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	2.388 ms	0 - 130 MB	NPU
WideResNet50	TFLITE	w8a8	Qualcomm® SA7255P	3.733 ms	0 - 49 MB	NPU
WideResNet50	TFLITE	w8a8	Qualcomm® SA8295P	2.579 ms	0 - 53 MB	NPU
WideResNet50	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	1.114 ms	0 - 54 MB	NPU
WideResNet50	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	2.747 ms	0 - 205 MB	NPU

License

• The license for the original implementation of WideResNet50 can be found here.

References

• Wide Residual Networks
• Source Model Implementation

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.