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	1.061 ms	1 - 75 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Elite Mobile	1.243 ms	0 - 54 MB	NPU
ResNet50	ONNX	float	Snapdragon® X2 Elite	0.98 ms	49 - 49 MB	NPU
ResNet50	ONNX	float	Snapdragon® X Elite	2.071 ms	49 - 49 MB	NPU
ResNet50	ONNX	float	Snapdragon® X Elite	2.071 ms	49 - 49 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.476 ms	0 - 83 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.989 ms	0 - 10 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS9075	3.177 ms	0 - 4 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.243 ms	0 - 54 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.549 ms	0 - 45 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite Mobile	0.581 ms	0 - 46 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X2 Elite	0.483 ms	25 - 25 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X Elite	0.954 ms	25 - 25 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X Elite	0.954 ms	25 - 25 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.677 ms	0 - 80 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS6490	32.0 ms	8 - 27 MB	CPU
ResNet50	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.894 ms	0 - 2 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS9075	0.969 ms	0 - 3 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCM6690	23.239 ms	6 - 14 MB	CPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.581 ms	0 - 46 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	17.798 ms	6 - 14 MB	CPU
ResNet50	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	17.798 ms	6 - 14 MB	CPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.083 ms	1 - 49 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Mobile	1.262 ms	1 - 49 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® X2 Elite	1.233 ms	1 - 1 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® X Elite	2.338 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	1.614 ms	0 - 77 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	10.717 ms	1 - 45 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	2.24 ms	1 - 93 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8775P	3.344 ms	1 - 47 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8775P	3.344 ms	1 - 47 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8775P	3.344 ms	1 - 47 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS9075	3.299 ms	3 - 5 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	3.752 ms	0 - 61 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA7255P	10.717 ms	1 - 45 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8295P	3.666 ms	0 - 28 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.262 ms	1 - 49 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.486 ms	0 - 46 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite Mobile	0.53 ms	0 - 43 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.515 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X Elite	0.955 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X Elite	0.955 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.678 ms	0 - 75 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS6490	2.984 ms	2 - 4 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	2.003 ms	0 - 42 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.897 ms	0 - 10 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8775P	1.088 ms	0 - 44 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8775P	1.088 ms	0 - 44 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8775P	1.088 ms	0 - 44 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS9075	0.969 ms	0 - 2 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCM6690	6.484 ms	0 - 162 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	1.194 ms	0 - 73 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA7255P	2.003 ms	0 - 42 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8295P	1.446 ms	0 - 40 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.53 ms	0 - 43 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.227 ms	0 - 50 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.227 ms	0 - 50 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.084 ms	0 - 71 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite Mobile	1.249 ms	0 - 72 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.581 ms	0 - 109 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8275 (Proxy)	10.619 ms	0 - 69 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8550 (Proxy)	2.234 ms	0 - 85 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8775P	3.34 ms	0 - 70 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8775P	3.34 ms	0 - 70 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8775P	3.34 ms	0 - 70 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS9075	3.337 ms	0 - 52 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8450 (Proxy)	3.73 ms	0 - 96 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA7255P	10.619 ms	0 - 69 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8295P	3.622 ms	0 - 50 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.249 ms	0 - 72 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.445 ms	0 - 46 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite Mobile	0.485 ms	0 - 45 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.578 ms	0 - 74 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS6490	2.727 ms	0 - 27 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	1.766 ms	0 - 42 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.763 ms	0 - 2 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8775P	0.966 ms	0 - 45 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8775P	0.966 ms	0 - 45 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8775P	0.966 ms	0 - 45 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS9075	0.82 ms	0 - 27 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCM6690	6.108 ms	0 - 162 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	1.049 ms	0 - 75 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA7255P	1.766 ms	0 - 42 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8295P	1.29 ms	0 - 40 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.485 ms	0 - 45 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.096 ms	0 - 49 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.096 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.