ResNet18: Optimized for Qualcomm Devices

ResNet18 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 ResNet18 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 ResNet18 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 ResNet18 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: 11.7M
• Model size (float): 44.6 MB
• Model size (w8a8): 11.3 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
ResNet18	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.557 ms	0 - 26 MB	NPU
ResNet18	ONNX	float	Snapdragon® X2 Elite	0.533 ms	23 - 23 MB	NPU
ResNet18	ONNX	float	Snapdragon® X Elite	1.175 ms	22 - 22 MB	NPU
ResNet18	ONNX	float	Snapdragon® 8 Gen 3 Mobile	0.773 ms	0 - 34 MB	NPU
ResNet18	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.068 ms	0 - 25 MB	NPU
ResNet18	ONNX	float	Qualcomm® QCS9075	1.796 ms	1 - 3 MB	NPU
ResNet18	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	0.626 ms	0 - 20 MB	NPU
ResNet18	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.339 ms	0 - 26 MB	NPU
ResNet18	ONNX	w8a8	Snapdragon® X2 Elite	0.309 ms	11 - 11 MB	NPU
ResNet18	ONNX	w8a8	Snapdragon® X Elite	0.642 ms	11 - 11 MB	NPU
ResNet18	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.403 ms	0 - 49 MB	NPU
ResNet18	ONNX	w8a8	Qualcomm® QCS6490	13.652 ms	7 - 22 MB	CPU
ResNet18	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.558 ms	0 - 2 MB	NPU
ResNet18	ONNX	w8a8	Qualcomm® QCS9075	0.64 ms	0 - 3 MB	NPU
ResNet18	ONNX	w8a8	Qualcomm® QCM6690	11.409 ms	8 - 15 MB	CPU
ResNet18	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.357 ms	0 - 22 MB	NPU
ResNet18	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	8.675 ms	6 - 13 MB	CPU
ResNet18	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.587 ms	1 - 28 MB	NPU
ResNet18	QNN_DLC	float	Snapdragon® X2 Elite	0.693 ms	1 - 1 MB	NPU
ResNet18	QNN_DLC	float	Snapdragon® X Elite	1.454 ms	1 - 1 MB	NPU
ResNet18	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	0.906 ms	0 - 38 MB	NPU
ResNet18	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	5.801 ms	1 - 24 MB	NPU
ResNet18	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	1.327 ms	1 - 103 MB	NPU
ResNet18	QNN_DLC	float	Qualcomm® SA8775P	1.95 ms	0 - 26 MB	NPU
ResNet18	QNN_DLC	float	Qualcomm® QCS9075	2.066 ms	3 - 5 MB	NPU
ResNet18	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	2.445 ms	0 - 34 MB	NPU
ResNet18	QNN_DLC	float	Qualcomm® SA7255P	5.801 ms	1 - 24 MB	NPU
ResNet18	QNN_DLC	float	Qualcomm® SA8295P	2.305 ms	0 - 19 MB	NPU
ResNet18	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	0.701 ms	0 - 23 MB	NPU
ResNet18	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.265 ms	0 - 25 MB	NPU
ResNet18	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.311 ms	0 - 0 MB	NPU
ResNet18	QNN_DLC	w8a8	Snapdragon® X Elite	0.623 ms	0 - 0 MB	NPU
ResNet18	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.391 ms	0 - 48 MB	NPU
ResNet18	QNN_DLC	w8a8	Qualcomm® QCS6490	1.817 ms	0 - 2 MB	NPU
ResNet18	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	1.274 ms	0 - 23 MB	NPU
ResNet18	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.512 ms	0 - 1 MB	NPU
ResNet18	QNN_DLC	w8a8	Qualcomm® SA8775P	0.7 ms	0 - 25 MB	NPU
ResNet18	QNN_DLC	w8a8	Qualcomm® QCS9075	0.626 ms	0 - 2 MB	NPU
ResNet18	QNN_DLC	w8a8	Qualcomm® QCM6690	3.227 ms	0 - 31 MB	NPU
ResNet18	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	0.741 ms	0 - 48 MB	NPU
ResNet18	QNN_DLC	w8a8	Qualcomm® SA7255P	1.274 ms	0 - 23 MB	NPU
ResNet18	QNN_DLC	w8a8	Qualcomm® SA8295P	0.897 ms	0 - 21 MB	NPU
ResNet18	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.294 ms	0 - 22 MB	NPU
ResNet18	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	0.685 ms	0 - 31 MB	NPU
ResNet18	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.585 ms	0 - 30 MB	NPU
ResNet18	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	0.91 ms	0 - 61 MB	NPU
ResNet18	TFLITE	float	Qualcomm® QCS8275 (Proxy)	5.751 ms	0 - 28 MB	NPU
ResNet18	TFLITE	float	Qualcomm® QCS8550 (Proxy)	1.338 ms	0 - 8 MB	NPU
ResNet18	TFLITE	float	Qualcomm® SA8775P	1.947 ms	0 - 30 MB	NPU
ResNet18	TFLITE	float	Qualcomm® QCS9075	2.035 ms	0 - 25 MB	NPU
ResNet18	TFLITE	float	Qualcomm® QCS8450 (Proxy)	2.47 ms	0 - 61 MB	NPU
ResNet18	TFLITE	float	Qualcomm® SA7255P	5.751 ms	0 - 28 MB	NPU
ResNet18	TFLITE	float	Qualcomm® SA8295P	2.276 ms	0 - 22 MB	NPU
ResNet18	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	0.703 ms	0 - 26 MB	NPU
ResNet18	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.218 ms	0 - 26 MB	NPU
ResNet18	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.299 ms	0 - 47 MB	NPU
ResNet18	TFLITE	w8a8	Qualcomm® QCS6490	1.528 ms	0 - 13 MB	NPU
ResNet18	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	1.012 ms	0 - 23 MB	NPU
ResNet18	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.385 ms	0 - 14 MB	NPU
ResNet18	TFLITE	w8a8	Qualcomm® SA8775P	0.572 ms	0 - 25 MB	NPU
ResNet18	TFLITE	w8a8	Qualcomm® QCS9075	0.48 ms	0 - 13 MB	NPU
ResNet18	TFLITE	w8a8	Qualcomm® QCM6690	2.794 ms	0 - 31 MB	NPU
ResNet18	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	0.611 ms	0 - 49 MB	NPU
ResNet18	TFLITE	w8a8	Qualcomm® SA7255P	1.012 ms	0 - 23 MB	NPU
ResNet18	TFLITE	w8a8	Qualcomm® SA8295P	0.742 ms	0 - 21 MB	NPU
ResNet18	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.237 ms	0 - 26 MB	NPU
ResNet18	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	0.55 ms	0 - 31 MB	NPU

License

• The license for the original implementation of ResNet18 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.