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 1.24.3	Download
ONNX	w8a8	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	Download
QNN_DLC	float	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 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® 8 Elite Gen 5 Mobile	1.052 ms	1 - 75 MB	NPU
ResNet50	ONNX	float	Snapdragon® X2 Elite	0.977 ms	49 - 49 MB	NPU
ResNet50	ONNX	float	Snapdragon® X Elite	2.074 ms	49 - 49 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.469 ms	0 - 85 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.991 ms	0 - 4 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS9075	3.189 ms	1 - 4 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.242 ms	0 - 54 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.554 ms	0 - 45 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X2 Elite	0.416 ms	25 - 25 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X Elite	0.964 ms	25 - 25 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.674 ms	0 - 82 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS6490	32.622 ms	8 - 27 MB	CPU
ResNet50	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.9 ms	0 - 30 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS9075	0.957 ms	0 - 3 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCM6690	23.357 ms	7 - 16 MB	CPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.583 ms	0 - 41 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	17.871 ms	6 - 14 MB	CPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.089 ms	1 - 49 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® X2 Elite	1.204 ms	1 - 1 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® X Elite	2.338 ms	1 - 1 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.59 ms	0 - 75 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	10.713 ms	1 - 46 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	2.21 ms	1 - 108 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8775P	3.355 ms	1 - 47 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS9075	3.302 ms	3 - 5 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	3.742 ms	0 - 61 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA7255P	10.713 ms	1 - 46 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8295P	3.662 ms	0 - 29 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.266 ms	1 - 51 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.485 ms	0 - 45 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.457 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X Elite	0.946 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.68 ms	0 - 75 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS6490	3.013 ms	2 - 4 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	2.007 ms	0 - 43 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.898 ms	0 - 116 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8775P	1.094 ms	0 - 44 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS9075	0.965 ms	2 - 4 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCM6690	6.495 ms	0 - 162 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	1.189 ms	0 - 73 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA7255P	2.007 ms	0 - 43 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8295P	1.455 ms	0 - 40 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.524 ms	0 - 40 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.221 ms	0 - 49 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.083 ms	0 - 72 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.567 ms	0 - 112 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8275 (Proxy)	10.65 ms	0 - 68 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8550 (Proxy)	2.226 ms	0 - 2 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8775P	3.333 ms	0 - 70 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS9075	3.332 ms	0 - 52 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8450 (Proxy)	3.731 ms	0 - 95 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA7255P	10.65 ms	0 - 68 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8295P	3.63 ms	0 - 50 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.252 ms	0 - 71 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.449 ms	0 - 46 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.578 ms	0 - 73 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS6490	2.768 ms	0 - 27 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	1.743 ms	0 - 42 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.753 ms	0 - 2 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8775P	0.957 ms	0 - 45 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS9075	0.833 ms	0 - 27 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCM6690	6.119 ms	0 - 162 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	1.049 ms	0 - 75 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA7255P	1.743 ms	0 - 42 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8295P	1.326 ms	0 - 39 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.495 ms	0 - 44 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.091 ms	0 - 49 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.