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 1.25.0	Download
ONNX	w8a8	Universal	QAIRT 2.42, ONNX Runtime 1.25.0	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	1.062 ms	1 - 73 MB	NPU
ResNet50	ONNX	float	Snapdragon® X2 Elite	0.983 ms	212 - 212 MB	NPU
ResNet50	ONNX	float	Snapdragon® X Elite	1.931 ms	149 - 149 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.474 ms	0 - 85 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS8550 (Proxy)	2.0 ms	0 - 69 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.243 ms	0 - 50 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS9075	3.177 ms	1 - 46 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS8750	1.243 ms	0 - 50 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS7181	1.931 ms	149 - 149 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.553 ms	0 - 50 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X2 Elite	0.41 ms	212 - 212 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.68 ms	0 - 85 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS6490	31.44 ms	8 - 27 MB	CPU
ResNet50	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.901 ms	0 - 40 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	17.721 ms	12 - 21 MB	CPU
ResNet50	ONNX	w8a8	Qualcomm® QCM6690	23.402 ms	7 - 15 MB	CPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.576 ms	0 - 49 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS9075	1.0 ms	0 - 45 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS7790	17.721 ms	12 - 21 MB	CPU
ResNet50	ONNX	w8a8	Qualcomm® QCS8750	0.576 ms	0 - 49 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.082 ms	1 - 50 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® X2 Elite	1.25 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	1.624 ms	1 - 78 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8275	10.731 ms	1 - 46 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	2.211 ms	1 - 2 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8775P	3.358 ms	0 - 47 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8650P	3.358 ms	0 - 47 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8255P	3.358 ms	0 - 47 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	3.745 ms	0 - 62 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8295P	3.669 ms	0 - 28 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.262 ms	0 - 50 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA7255P	10.731 ms	1 - 46 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS9075	3.325 ms	1 - 3 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8750	1.262 ms	0 - 50 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS7181	2.338 ms	1 - 1 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.477 ms	0 - 47 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.477 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X Elite	0.932 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.688 ms	0 - 78 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS6490	2.922 ms	0 - 2 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8275	2.058 ms	0 - 45 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.902 ms	0 - 11 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8775P	1.092 ms	0 - 46 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8650P	1.092 ms	0 - 46 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8255P	1.092 ms	0 - 46 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA7255P	2.058 ms	0 - 45 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.222 ms	0 - 51 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8295P	1.458 ms	0 - 42 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCM6690	6.621 ms	0 - 164 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.536 ms	0 - 47 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS9075	0.966 ms	2 - 4 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	1.193 ms	0 - 75 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS7790	1.222 ms	0 - 51 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8750	0.536 ms	0 - 47 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS7181	0.932 ms	0 - 0 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.08 ms	0 - 71 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.59 ms	0 - 113 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8550 (Proxy)	2.226 ms	0 - 2 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8775P	3.352 ms	0 - 70 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8650P	3.352 ms	0 - 70 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8255P	3.352 ms	0 - 70 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8450 (Proxy)	3.735 ms	0 - 96 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8295P	3.647 ms	0 - 50 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.251 ms	0 - 70 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS9075	3.331 ms	0 - 52 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8750	1.251 ms	0 - 70 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.445 ms	0 - 46 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.576 ms	0 - 78 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS6490	2.767 ms	0 - 27 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8275	1.745 ms	0 - 42 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.759 ms	0 - 11 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8775P	0.963 ms	0 - 45 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8650P	0.963 ms	0 - 45 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8255P	0.963 ms	0 - 45 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA7255P	1.745 ms	0 - 42 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.096 ms	0 - 49 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8295P	1.285 ms	0 - 39 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCM6690	6.229 ms	0 - 161 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.482 ms	0 - 40 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS9075	0.818 ms	0 - 27 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	1.052 ms	0 - 75 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS7790	1.096 ms	0 - 49 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8750	0.482 ms	0 - 40 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.