RegNet: Optimized for Qualcomm Devices

RegNet 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 RegNet 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 2.45, ONNX Runtime 1.27.1	Download
ONNX	w8a8	Universal	QAIRT 2.45, ONNX Runtime 1.27.1	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 RegNet 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 RegNet 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): 58.3 MB
Model size (w8a8): 15.4 MB
Number of parameters: 15.3M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
RegNet	ONNX	float	Snapdragon® X2 Elite	0.902 ms	2 - 2 MB	NPU
RegNet	ONNX	float	Snapdragon® X Elite	1.779 ms	39 - 39 MB	NPU
RegNet	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.214 ms	0 - 130 MB	NPU
RegNet	ONNX	float	Snapdragon® 8 Gen 1 Mobile	2.736 ms	0 - 116 MB	NPU
RegNet	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	3.03 ms	0 - 4 MB	NPU
RegNet	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.742 ms	0 - 57 MB	NPU
RegNet	ONNX	float	Qualcomm® QCS8450	2.736 ms	0 - 116 MB	NPU
RegNet	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	2.813 ms	0 - 4 MB	NPU
RegNet	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	1.779 ms	39 - 39 MB	NPU
RegNet	ONNX	float	Qualcomm® Dragonwing™ Q-8750	0.998 ms	0 - 78 MB	NPU
RegNet	ONNX	float	Snapdragon® 8 Elite Mobile	0.998 ms	0 - 78 MB	NPU
RegNet	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.839 ms	0 - 77 MB	NPU
RegNet	ONNX	w8a8	Snapdragon® X2 Elite	0.472 ms	1 - 1 MB	NPU
RegNet	ONNX	w8a8	Snapdragon® X Elite	0.928 ms	20 - 20 MB	NPU
RegNet	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.646 ms	0 - 135 MB	NPU
RegNet	ONNX	w8a8	Snapdragon® 8 Gen 1 Mobile	1.306 ms	0 - 134 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® Dragonwing™ QCS6490	3.965 ms	0 - 3 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® Dragonwing™ IQ-8275	1.029 ms	0 - 4 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® Dragonwing™ QCS8550 (Proxy)	0.894 ms	0 - 163 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® QCS8450	1.306 ms	0 - 134 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® Dragonwing™ IQ-9075	1.167 ms	0 - 3 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® Dragonwing™ IQ-X7181	0.928 ms	20 - 20 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® Dragonwing™ Q-6690	7.221 ms	0 - 206 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® Dragonwing™ Q-7790	1.135 ms	0 - 87 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® Dragonwing™ Q-8750	0.549 ms	0 - 89 MB	NPU
RegNet	ONNX	w8a8	Snapdragon® 8 Elite Mobile	0.549 ms	0 - 89 MB	NPU
RegNet	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.515 ms	0 - 88 MB	NPU
RegNet	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	1.135 ms	0 - 87 MB	NPU
RegNet	QNN_DLC	float	Snapdragon® X2 Elite	1.203 ms	1 - 1 MB	NPU
RegNet	QNN_DLC	float	Snapdragon® X Elite	2.264 ms	1 - 1 MB	NPU
RegNet	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.41 ms	0 - 127 MB	NPU
RegNet	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	3.5 ms	0 - 116 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	3.182 ms	1 - 4 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	9.855 ms	1 - 72 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	2.052 ms	1 - 2 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® SA8775P	3.286 ms	1 - 74 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® SA8650P	3.286 ms	1 - 74 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® SA8255P	3.286 ms	1 - 74 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® QCS8450	3.5 ms	0 - 116 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	3.059 ms	1 - 3 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	2.264 ms	1 - 1 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	1.109 ms	1 - 75 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® SA7255P	9.855 ms	1 - 72 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® SA8295P	3.485 ms	0 - 61 MB	NPU
RegNet	QNN_DLC	float	Snapdragon® 8 Elite Mobile	1.109 ms	1 - 75 MB	NPU
RegNet	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.906 ms	1 - 77 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.571 ms	0 - 0 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® X Elite	1.059 ms	0 - 0 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.654 ms	0 - 113 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® 8 Gen 1 Mobile	1.308 ms	0 - 111 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® Dragonwing™ QCS6490	4.089 ms	0 - 2 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-8275	1.0 ms	0 - 3 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-8275	2.282 ms	0 - 80 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® Dragonwing™ QCS8550 (Proxy)	0.889 ms	0 - 2 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® SA8775P	1.288 ms	0 - 80 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® SA8650P	1.288 ms	0 - 80 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® SA8255P	1.288 ms	0 - 80 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® QCS8450	1.308 ms	0 - 111 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-9075	1.158 ms	0 - 2 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® Dragonwing™ IQ-X7181	1.059 ms	0 - 0 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® Dragonwing™ Q-6690	7.135 ms	0 - 199 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® Dragonwing™ Q-7790	1.177 ms	0 - 80 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® Dragonwing™ Q-8750	0.481 ms	0 - 78 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® SA7255P	2.282 ms	0 - 80 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® SA8295P	1.605 ms	0 - 78 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® 8 Elite Mobile	0.481 ms	0 - 78 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.438 ms	0 - 83 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.177 ms	0 - 80 MB	NPU
RegNet	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.413 ms	0 - 150 MB	NPU
RegNet	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	3.449 ms	0 - 139 MB	NPU
RegNet	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	3.176 ms	0 - 42 MB	NPU
RegNet	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	9.856 ms	0 - 90 MB	NPU
RegNet	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	2.021 ms	0 - 5 MB	NPU
RegNet	TFLITE	float	Qualcomm® SA8775P	3.316 ms	0 - 91 MB	NPU
RegNet	TFLITE	float	Qualcomm® SA8650P	3.316 ms	0 - 91 MB	NPU
RegNet	TFLITE	float	Qualcomm® SA8255P	3.316 ms	0 - 91 MB	NPU
RegNet	TFLITE	float	Qualcomm® QCS8450	3.449 ms	0 - 139 MB	NPU
RegNet	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	3.053 ms	0 - 42 MB	NPU
RegNet	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	1.103 ms	0 - 92 MB	NPU
RegNet	TFLITE	float	Qualcomm® SA7255P	9.856 ms	0 - 90 MB	NPU
RegNet	TFLITE	float	Qualcomm® SA8295P	3.48 ms	0 - 72 MB	NPU
RegNet	TFLITE	float	Snapdragon® 8 Elite Mobile	1.103 ms	0 - 92 MB	NPU
RegNet	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.898 ms	0 - 86 MB	NPU
RegNet	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.525 ms	0 - 120 MB	NPU
RegNet	TFLITE	w8a8	Snapdragon® 8 Gen 1 Mobile	1.101 ms	0 - 116 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® Dragonwing™ QCS6490	3.879 ms	0 - 22 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® Dragonwing™ IQ-8275	0.821 ms	0 - 23 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® Dragonwing™ IQ-8275	1.995 ms	0 - 76 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® Dragonwing™ QCS8550 (Proxy)	0.717 ms	0 - 3 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® SA8775P	1.108 ms	0 - 79 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® SA8650P	1.108 ms	0 - 79 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® SA8255P	1.108 ms	0 - 79 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® QCS8450	1.101 ms	0 - 116 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® Dragonwing™ IQ-9075	0.931 ms	0 - 22 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® Dragonwing™ Q-6690	6.64 ms	0 - 190 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® Dragonwing™ Q-7790	0.981 ms	0 - 73 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® Dragonwing™ Q-8750	0.423 ms	0 - 73 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® SA7255P	1.995 ms	0 - 76 MB	NPU
RegNet	TFLITE	w8a8	Snapdragon® 8 Elite Mobile	0.423 ms	0 - 73 MB	NPU
RegNet	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.39 ms	0 - 78 MB	NPU
RegNet	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	0.981 ms	0 - 73 MB	NPU

License

The license for the original implementation of RegNet can be found here.

References

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.

Downloads last month: -; Downloads are not tracked for this model. How to track

Paper for qualcomm/RegNet

Designing Network Design Spaces

Paper • 2003.13678 • Published Mar 30, 2020 • 2