ResNet101: Optimized for Qualcomm Devices

ResNet101 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 ResNet101 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 ResNet101 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 ResNet101 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: 44.5M
• Model size (float): 170 MB
• Model size (w8a8): 43.9 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
ResNet101	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	1.604 ms	1 - 71 MB	NPU
ResNet101	ONNX	float	Snapdragon® X2 Elite	1.628 ms	212 - 212 MB	NPU
ResNet101	ONNX	float	Snapdragon® X Elite	3.074 ms	148 - 148 MB	NPU
ResNet101	ONNX	float	Snapdragon® 8 Gen 3 Mobile	2.236 ms	0 - 126 MB	NPU
ResNet101	ONNX	float	Qualcomm® QCS8550 (Proxy)	3.078 ms	0 - 108 MB	NPU
ResNet101	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.855 ms	0 - 70 MB	NPU
ResNet101	ONNX	float	Qualcomm® QCS9075	5.156 ms	1 - 46 MB	NPU
ResNet101	ONNX	float	Qualcomm® QCS8750	1.855 ms	0 - 70 MB	NPU
ResNet101	ONNX	float	Qualcomm® QCS7181	3.074 ms	148 - 148 MB	NPU
ResNet101	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.804 ms	0 - 84 MB	NPU
ResNet101	ONNX	w8a8	Snapdragon® X2 Elite	0.56 ms	212 - 212 MB	NPU
ResNet101	ONNX	w8a8	Snapdragon® X Elite	1.157 ms	149 - 149 MB	NPU
ResNet101	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.974 ms	0 - 147 MB	NPU
ResNet101	ONNX	w8a8	Qualcomm® QCS6490	56.406 ms	9 - 59 MB	CPU
ResNet101	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	1.284 ms	0 - 15 MB	NPU
ResNet101	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	39.465 ms	8 - 19 MB	CPU
ResNet101	ONNX	w8a8	Qualcomm® QCM6690	42.163 ms	15 - 26 MB	CPU
ResNet101	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.839 ms	0 - 82 MB	NPU
ResNet101	ONNX	w8a8	Qualcomm® QCS9075	1.335 ms	0 - 45 MB	NPU
ResNet101	ONNX	w8a8	Qualcomm® QCS7790	39.465 ms	8 - 19 MB	CPU
ResNet101	ONNX	w8a8	Qualcomm® QCS8750	0.839 ms	0 - 82 MB	NPU
ResNet101	ONNX	w8a8	Qualcomm® QCS7181	1.157 ms	149 - 149 MB	NPU
ResNet101	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.638 ms	2 - 70 MB	NPU
ResNet101	QNN_DLC	float	Snapdragon® X2 Elite	1.943 ms	1 - 1 MB	NPU
ResNet101	QNN_DLC	float	Snapdragon® X Elite	3.56 ms	1 - 1 MB	NPU
ResNet101	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	2.42 ms	0 - 121 MB	NPU
ResNet101	QNN_DLC	float	Qualcomm® QCS8275	18.328 ms	1 - 66 MB	NPU
ResNet101	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	3.312 ms	1 - 261 MB	NPU
ResNet101	QNN_DLC	float	Qualcomm® SA8775P	5.42 ms	1 - 67 MB	NPU
ResNet101	QNN_DLC	float	Qualcomm® SA8650P	5.42 ms	1 - 67 MB	NPU
ResNet101	QNN_DLC	float	Qualcomm® SA8255P	5.42 ms	1 - 67 MB	NPU
ResNet101	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	5.941 ms	0 - 90 MB	NPU
ResNet101	QNN_DLC	float	Qualcomm® SA8295P	5.649 ms	0 - 39 MB	NPU
ResNet101	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.942 ms	0 - 66 MB	NPU
ResNet101	QNN_DLC	float	Qualcomm® SA7255P	18.328 ms	1 - 66 MB	NPU
ResNet101	QNN_DLC	float	Qualcomm® QCS9075	5.284 ms	1 - 3 MB	NPU
ResNet101	QNN_DLC	float	Qualcomm® QCS8750	1.942 ms	0 - 66 MB	NPU
ResNet101	QNN_DLC	float	Qualcomm® QCS7181	3.56 ms	1 - 1 MB	NPU
ResNet101	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.742 ms	0 - 79 MB	NPU
ResNet101	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.628 ms	0 - 0 MB	NPU
ResNet101	QNN_DLC	w8a8	Snapdragon® X Elite	1.249 ms	0 - 0 MB	NPU
ResNet101	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.961 ms	0 - 125 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® QCS6490	4.698 ms	0 - 2 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® QCS8275	3.027 ms	0 - 74 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	1.288 ms	0 - 2 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® SA8775P	1.511 ms	0 - 77 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® SA8650P	1.511 ms	0 - 77 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® SA8255P	1.511 ms	0 - 77 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® SA7255P	3.027 ms	0 - 74 MB	NPU
ResNet101	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.8 ms	0 - 84 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® SA8295P	1.967 ms	0 - 73 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® QCM6690	11.823 ms	0 - 198 MB	NPU
ResNet101	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.787 ms	0 - 77 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® QCS9075	1.327 ms	0 - 2 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	1.785 ms	0 - 126 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® QCS7790	1.8 ms	0 - 84 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® QCS8750	0.787 ms	0 - 77 MB	NPU
ResNet101	QNN_DLC	w8a8	Qualcomm® QCS7181	1.249 ms	0 - 0 MB	NPU
ResNet101	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.623 ms	0 - 111 MB	NPU
ResNet101	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	2.378 ms	0 - 173 MB	NPU
ResNet101	TFLITE	float	Qualcomm® QCS8275	18.258 ms	0 - 108 MB	NPU
ResNet101	TFLITE	float	Qualcomm® QCS8550 (Proxy)	3.328 ms	0 - 2 MB	NPU
ResNet101	TFLITE	float	Qualcomm® SA8775P	5.412 ms	0 - 112 MB	NPU
ResNet101	TFLITE	float	Qualcomm® SA8650P	5.412 ms	0 - 112 MB	NPU
ResNet101	TFLITE	float	Qualcomm® SA8255P	5.412 ms	0 - 112 MB	NPU
ResNet101	TFLITE	float	Qualcomm® QCS8450 (Proxy)	5.905 ms	0 - 145 MB	NPU
ResNet101	TFLITE	float	Qualcomm® SA8295P	5.565 ms	0 - 83 MB	NPU
ResNet101	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.916 ms	0 - 111 MB	NPU
ResNet101	TFLITE	float	Qualcomm® SA7255P	18.258 ms	0 - 108 MB	NPU
ResNet101	TFLITE	float	Qualcomm® QCS9075	5.291 ms	0 - 88 MB	NPU
ResNet101	TFLITE	float	Qualcomm® QCS8750	1.916 ms	0 - 111 MB	NPU
ResNet101	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.7 ms	0 - 74 MB	NPU
ResNet101	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.862 ms	0 - 133 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® QCS6490	4.353 ms	0 - 45 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® QCS8275	2.755 ms	0 - 71 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	1.135 ms	0 - 3 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® SA8775P	1.396 ms	0 - 74 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® SA8650P	1.396 ms	0 - 74 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® SA8255P	1.396 ms	0 - 74 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® SA7255P	2.755 ms	0 - 71 MB	NPU
ResNet101	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.661 ms	0 - 78 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® SA8295P	1.801 ms	0 - 75 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® QCM6690	12.422 ms	0 - 194 MB	NPU
ResNet101	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.742 ms	0 - 72 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® QCS9075	1.171 ms	0 - 45 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	1.645 ms	0 - 136 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® QCS7790	1.661 ms	0 - 78 MB	NPU
ResNet101	TFLITE	w8a8	Qualcomm® QCS8750	0.742 ms	0 - 72 MB	NPU

License

• The license for the original implementation of ResNet101 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.