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 2.50, ONNX Runtime 1.30.0	Download
ONNX	w8a8	Universal	QAIRT 2.50, ONNX Runtime 1.30.0	Download
QNN_DLC	float	Universal	QAIRT 2.50	Download
QNN_DLC	w8a8	Universal	QAIRT 2.50	Download
TFLITE	float	Universal	QAIRT 2.50	Download
TFLITE	w8a8	Universal	QAIRT 2.50	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:

• Input resolution: 224x224
• Model checkpoint: Imagenet
• Model size (float): 97.4 MB
• Model size (w8a8): 25.1 MB
• Number of parameters: 25.5M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
ResNet50	ONNX	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	1.07 ms	0 - 53 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.224 ms	0 - 50 MB	NPU
ResNet50	ONNX	float	Snapdragon® X2 Elite	0.964 ms	2 - 2 MB	NPU
ResNet50	ONNX	float	Snapdragon® X Elite	1.922 ms	50 - 50 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.434 ms	0 - 82 MB	NPU
ResNet50	ONNX	float	Snapdragon® 8 Gen 1 Mobile	2.889 ms	0 - 67 MB	NPU
ResNet50	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	3.3 ms	0 - 5 MB	NPU
ResNet50	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.981 ms	0 - 71 MB	NPU
ResNet50	ONNX	float	Qualcomm® QCS8450	2.889 ms	0 - 67 MB	NPU
ResNet50	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	3.0 ms	1 - 4 MB	NPU
ResNet50	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	1.922 ms	50 - 50 MB	NPU
ResNet50	ONNX	float	Qualcomm® Dragonwing™ Q-8750	1.224 ms	0 - 50 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.559 ms	0 - 58 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.578 ms	0 - 53 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X2 Elite	0.409 ms	1 - 1 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® X Elite	0.843 ms	25 - 25 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.674 ms	0 - 89 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 8 Gen 1 Mobile	1.055 ms	0 - 90 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® Dragonwing™ QCS6490	2.64 ms	0 - 3 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® Dragonwing™ IQ-8275	0.977 ms	0 - 4 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® Dragonwing™ QCS8550 (Proxy)	0.91 ms	0 - 30 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® QCS8450	1.055 ms	0 - 90 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® Dragonwing™ IQ-9075	1.018 ms	0 - 3 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® Dragonwing™ IQ-X7181	0.843 ms	25 - 25 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® Dragonwing™ Q-6690	6.219 ms	0 - 170 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® Dragonwing™ Q-7790	1.134 ms	0 - 59 MB	NPU
ResNet50	ONNX	w8a8	Qualcomm® Dragonwing™ Q-8750	0.578 ms	0 - 53 MB	NPU
ResNet50	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	1.134 ms	0 - 59 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	1.096 ms	1 - 53 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.281 ms	0 - 47 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® X2 Elite	1.225 ms	1 - 1 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® X Elite	2.348 ms	1 - 1 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.617 ms	0 - 79 MB	NPU
ResNet50	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	3.475 ms	0 - 67 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	3.36 ms	1 - 4 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	2.245 ms	1 - 2 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8775P	3.316 ms	1 - 50 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8650P	3.316 ms	1 - 50 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8255P	3.316 ms	1 - 50 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® QCS8450	3.475 ms	0 - 67 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	3.168 ms	3 - 5 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	2.348 ms	1 - 1 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	1.281 ms	0 - 47 MB	NPU
ResNet50	QNN_DLC	float	Qualcomm® SA8295P	3.548 ms	0 - 32 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.486 ms	0 - 48 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.534 ms	0 - 51 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.444 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® X Elite	0.939 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.657 ms	0 - 77 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 8 Gen 1 Mobile	1.054 ms	0 - 82 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® Dragonwing™ QCS6490	2.96 ms	2 - 4 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-8275	0.943 ms	0 - 3 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® Dragonwing™ QCS8550 (Proxy)	0.901 ms	0 - 2 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8775P	1.086 ms	0 - 52 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8650P	1.086 ms	0 - 52 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8255P	1.086 ms	0 - 52 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® QCS8450	1.054 ms	0 - 82 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-9075	0.965 ms	0 - 2 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-X7181	0.939 ms	0 - 0 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® Dragonwing™ Q-6690	6.443 ms	0 - 161 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® Dragonwing™ Q-7790	1.222 ms	0 - 55 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® Dragonwing™ Q-8750	0.534 ms	0 - 51 MB	NPU
ResNet50	QNN_DLC	w8a8	Qualcomm® SA8295P	1.309 ms	0 - 47 MB	NPU
ResNet50	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.222 ms	0 - 55 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	1.107 ms	0 - 72 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.257 ms	0 - 69 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.592 ms	0 - 118 MB	NPU
ResNet50	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	3.471 ms	0 - 103 MB	NPU
ResNet50	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	3.334 ms	0 - 53 MB	NPU
ResNet50	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	2.203 ms	0 - 3 MB	NPU
ResNet50	TFLITE	float	Qualcomm® QCS8450	3.471 ms	0 - 103 MB	NPU
ResNet50	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	3.132 ms	0 - 52 MB	NPU
ResNet50	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	1.257 ms	0 - 69 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA7255P	10.595 ms	0 - 71 MB	NPU
ResNet50	TFLITE	float	Qualcomm® SA8295P	3.501 ms	0 - 54 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.453 ms	0 - 46 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.477 ms	0 - 43 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.585 ms	0 - 78 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 8 Gen 1 Mobile	0.921 ms	0 - 86 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® Dragonwing™ QCS6490	2.662 ms	0 - 27 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® Dragonwing™ IQ-8275	0.835 ms	0 - 28 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® Dragonwing™ QCS8550 (Proxy)	0.752 ms	0 - 2 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® QCS8450	0.921 ms	0 - 86 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® Dragonwing™ IQ-9075	0.839 ms	0 - 27 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® Dragonwing™ Q-6690	5.867 ms	0 - 157 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® Dragonwing™ Q-7790	1.08 ms	0 - 53 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® Dragonwing™ Q-8750	0.477 ms	0 - 43 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA7255P	1.759 ms	0 - 47 MB	NPU
ResNet50	TFLITE	w8a8	Qualcomm® SA8295P	1.154 ms	0 - 45 MB	NPU
ResNet50	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.08 ms	0 - 53 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.