ResNet50: Optimized for Qualcomm Devices

ResNet50 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 ResNet50 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 ResNet50 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 ResNet50 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: 25.5M
• Model size (float): 97.4 MB
• Model size (w8a8): 25.1 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
ResNet50	ONNX	float	Snapdragon® X Elite	2.093 ms	49 - 49 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.487 ms	0 - 80 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.993 ms	1 - 3 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS9075	3.173 ms	0 - 4 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.242 ms	0 - 54 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	1.062 ms	1 - 75 MB	NPU
ResNet50	ONNX	float	Snapdragon® X2 Elite	0.99 ms	49 - 49 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X Elite	0.972 ms	25 - 25 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.683 ms	0 - 78 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS6490	31.67 ms	9 - 28 MB	CPU
ResNet50	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.9 ms	0 - 30 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS9075	0.973 ms	0 - 3 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCM6690	23.437 ms	6 - 15 MB	CPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.597 ms	0 - 48 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	17.748 ms	11 - 19 MB	CPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.555 ms	0 - 44 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X2 Elite	0.423 ms	25 - 25 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® X Elite	2.328 ms	1 - 1 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.598 ms	0 - 77 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	10.61 ms	1 - 46 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	2.192 ms	1 - 2 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8775P	3.365 ms	1 - 47 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS9075	3.321 ms	1 - 3 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	3.759 ms	0 - 62 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA7255P	10.61 ms	1 - 46 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8295P	3.663 ms	0 - 30 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.259 ms	0 - 48 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.083 ms	1 - 51 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® X2 Elite	1.265 ms	1 - 1 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X Elite	0.963 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.679 ms	0 - 71 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS6490	2.981 ms	2 - 4 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	1.983 ms	0 - 41 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.897 ms	0 - 35 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8775P	1.106 ms	0 - 44 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS9075	0.973 ms	2 - 4 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCM6690	6.5 ms	0 - 161 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	1.193 ms	0 - 72 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA7255P	1.983 ms	0 - 41 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8295P	1.427 ms	0 - 38 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.532 ms	0 - 42 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.236 ms	0 - 48 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.485 ms	0 - 42 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.471 ms	0 - 0 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.579 ms	0 - 118 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8275 (Proxy)	10.521 ms	0 - 80 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8550 (Proxy)	2.209 ms	0 - 2 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8775P	3.372 ms	0 - 81 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS9075	3.364 ms	0 - 52 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8450 (Proxy)	3.738 ms	0 - 107 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA7255P	10.521 ms	0 - 80 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8295P	3.635 ms	0 - 67 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.263 ms	0 - 82 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.085 ms	0 - 83 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.592 ms	0 - 73 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS6490	2.729 ms	0 - 27 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	1.748 ms	0 - 40 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.762 ms	0 - 7 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8775P	0.983 ms	0 - 43 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS9075	0.825 ms	0 - 27 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCM6690	6.238 ms	0 - 159 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	1.039 ms	0 - 73 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA7255P	1.748 ms	0 - 40 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8295P	1.284 ms	0 - 38 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.509 ms	0 - 42 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.083 ms	0 - 46 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.444 ms	0 - 42 MB	NPU

License

• The license for the original implementation of ResNet50 can be found here.

References

• Deep Residual Learning for Image Recognition
• 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.