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 2.43	Download
QNN_DLC	w8a8	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.19.1	Download
TFLITE	w8a8	Universal	QAIRT 2.43, TFLite 2.19.1	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	1.056 ms	1 - 75 MB	NPU
ResNet50	ONNX	float	Snapdragon® X2 Elite	0.981 ms	49 - 49 MB	NPU
ResNet50	ONNX	float	Snapdragon® X Elite	2.087 ms	49 - 49 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.485 ms	0 - 85 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.999 ms	0 - 33 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS9075	3.175 ms	0 - 4 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.275 ms	0 - 54 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.554 ms	0 - 46 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X2 Elite	0.416 ms	25 - 25 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X Elite	0.965 ms	25 - 25 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.681 ms	0 - 82 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS6490	32.093 ms	10 - 29 MB	CPU
ResNet50	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.9 ms	0 - 31 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS9075	0.972 ms	0 - 3 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCM6690	23.262 ms	5 - 13 MB	CPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.595 ms	0 - 48 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	18.001 ms	8 - 16 MB	CPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.083 ms	1 - 52 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® X2 Elite	1.273 ms	1 - 1 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® X Elite	2.314 ms	1 - 1 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.625 ms	1 - 81 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	10.627 ms	1 - 47 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	2.189 ms	1 - 143 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8775P	3.369 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)	3.75 ms	0 - 62 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA7255P	10.627 ms	1 - 47 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8295P	3.686 ms	0 - 30 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.263 ms	0 - 49 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.479 ms	0 - 42 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.481 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X Elite	0.967 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.686 ms	0 - 70 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS6490	2.942 ms	2 - 4 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	2.017 ms	0 - 41 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.897 ms	0 - 112 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8775P	1.111 ms	0 - 44 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS9075	0.973 ms	0 - 2 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCM6690	6.469 ms	0 - 162 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	1.19 ms	0 - 72 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA7255P	2.017 ms	0 - 41 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8295P	1.441 ms	0 - 38 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.523 ms	0 - 38 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.234 ms	0 - 48 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.083 ms	0 - 83 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.576 ms	0 - 121 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8275 (Proxy)	10.524 ms	0 - 80 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8550 (Proxy)	2.202 ms	0 - 2 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8775P	3.373 ms	0 - 81 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS9075	3.354 ms	0 - 52 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8450 (Proxy)	3.732 ms	0 - 107 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA7255P	10.524 ms	0 - 80 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8295P	3.633 ms	0 - 61 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.279 ms	0 - 85 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.447 ms	0 - 43 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.591 ms	0 - 75 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS6490	2.705 ms	0 - 26 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	1.76 ms	0 - 42 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.761 ms	0 - 9 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8775P	0.991 ms	0 - 45 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS9075	0.827 ms	0 - 27 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCM6690	6.24 ms	0 - 165 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	1.055 ms	0 - 77 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA7255P	1.76 ms	0 - 42 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8295P	1.291 ms	0 - 39 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.479 ms	0 - 40 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.083 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.