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 2.42, ONNX Runtime 1.24.1	Download
ONNX	w8a8	Universal	QAIRT 2.42, ONNX Runtime 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
QNN_DLC	w8a8	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.17.0	Download
TFLITE	w8a8	Universal	QAIRT 2.43, TFLite 2.17.0	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:

• Model checkpoint: Imagenet
• Input resolution: 224x224
• Number of parameters: 15.3M
• Model size (float): 58.3 MB
• Model size (w8a8): 15.4 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
RegNet	ONNX	float	Snapdragon® X2 Elite	0.906 ms	39 - 39 MB	NPU
RegNet	ONNX	float	Snapdragon® X Elite	1.958 ms	39 - 39 MB	NPU
RegNet	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.24 ms	0 - 141 MB	NPU
RegNet	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.751 ms	0 - 54 MB	NPU
RegNet	ONNX	float	Qualcomm® QCS9075	2.856 ms	0 - 4 MB	NPU
RegNet	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	1.013 ms	0 - 88 MB	NPU
RegNet	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.836 ms	1 - 88 MB	NPU
RegNet	ONNX	w8a8	Snapdragon® X2 Elite	0.518 ms	20 - 20 MB	NPU
RegNet	ONNX	w8a8	Snapdragon® X Elite	1.103 ms	20 - 20 MB	NPU
RegNet	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.652 ms	0 - 133 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® QCS6490	28.232 ms	8 - 20 MB	CPU
RegNet	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.895 ms	0 - 30 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® QCS9075	1.077 ms	0 - 3 MB	NPU
RegNet	ONNX	w8a8	Qualcomm® QCM6690	18.271 ms	7 - 16 MB	CPU
RegNet	ONNX	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.545 ms	0 - 79 MB	NPU
RegNet	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	13.94 ms	8 - 17 MB	CPU
RegNet	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.511 ms	0 - 89 MB	NPU
RegNet	QNN_DLC	float	Snapdragon® X2 Elite	1.227 ms	1 - 1 MB	NPU
RegNet	QNN_DLC	float	Snapdragon® X Elite	2.327 ms	1 - 1 MB	NPU
RegNet	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.399 ms	0 - 127 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	9.918 ms	1 - 77 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	2.079 ms	1 - 2 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® SA8775P	3.286 ms	1 - 79 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® QCS9075	3.068 ms	1 - 3 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	3.483 ms	0 - 117 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® SA7255P	9.918 ms	1 - 77 MB	NPU
RegNet	QNN_DLC	float	Qualcomm® SA8295P	3.528 ms	0 - 65 MB	NPU
RegNet	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	1.121 ms	1 - 78 MB	NPU
RegNet	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.9 ms	1 - 81 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.598 ms	0 - 0 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® X Elite	1.073 ms	0 - 0 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.643 ms	0 - 110 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® QCS6490	2.693 ms	0 - 2 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® QCS8275 (Proxy)	2.288 ms	0 - 76 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.888 ms	0 - 2 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® SA8775P	4.983 ms	0 - 76 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® QCS9075	1.081 ms	2 - 4 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® QCM6690	7.096 ms	0 - 197 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® QCS8450 (Proxy)	1.366 ms	0 - 110 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® SA7255P	2.288 ms	0 - 76 MB	NPU
RegNet	QNN_DLC	w8a8	Qualcomm® SA8295P	1.636 ms	0 - 75 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.487 ms	0 - 76 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.148 ms	0 - 77 MB	NPU
RegNet	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.441 ms	0 - 80 MB	NPU
RegNet	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.408 ms	0 - 160 MB	NPU
RegNet	TFLITE	float	Qualcomm® QCS8275 (Proxy)	9.901 ms	0 - 101 MB	NPU
RegNet	TFLITE	float	Qualcomm® QCS8550 (Proxy)	2.047 ms	0 - 2 MB	NPU
RegNet	TFLITE	float	Qualcomm® SA8775P	14.216 ms	0 - 100 MB	NPU
RegNet	TFLITE	float	Qualcomm® QCS9075	3.056 ms	0 - 42 MB	NPU
RegNet	TFLITE	float	Qualcomm® QCS8450 (Proxy)	3.448 ms	0 - 142 MB	NPU
RegNet	TFLITE	float	Qualcomm® SA7255P	9.901 ms	0 - 101 MB	NPU
RegNet	TFLITE	float	Qualcomm® SA8295P	3.497 ms	0 - 83 MB	NPU
RegNet	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.109 ms	0 - 104 MB	NPU
RegNet	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.895 ms	0 - 104 MB	NPU
RegNet	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.54 ms	0 - 119 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® QCS6490	2.257 ms	0 - 21 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® QCS8275 (Proxy)	1.984 ms	0 - 74 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.731 ms	0 - 3 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® SA8775P	1.14 ms	0 - 76 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® QCS9075	0.89 ms	0 - 22 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® QCM6690	6.644 ms	0 - 190 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® QCS8450 (Proxy)	1.099 ms	0 - 117 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® SA7255P	1.984 ms	0 - 74 MB	NPU
RegNet	TFLITE	w8a8	Qualcomm® SA8295P	1.412 ms	0 - 71 MB	NPU
RegNet	TFLITE	w8a8	Snapdragon® 8 Elite For Galaxy Mobile	0.423 ms	0 - 79 MB	NPU
RegNet	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	0.988 ms	0 - 71 MB	NPU
RegNet	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.392 ms	0 - 78 MB	NPU

License

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

References

• Designing Network Design Spaces
• 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.