EfficientNet-B0: Optimized for Qualcomm Devices

EfficientNetB0 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 EfficientNet-B0 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 1.25.0	Download
ONNX	w8a16	Universal	QAIRT 2.45, ONNX Runtime 1.25.0	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
QNN_DLC	w8a16	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit EfficientNet-B0 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 EfficientNet-B0 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: 5.27M
Model size (float): 20.1 MB
Model size (w8a16): 6.99 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
EfficientNet-B0	ONNX	float	Snapdragon® X2 Elite	0.651 ms	212 - 212 MB	NPU
EfficientNet-B0	ONNX	float	Snapdragon® X Elite	1.316 ms	149 - 149 MB	NPU
EfficientNet-B0	ONNX	float	Snapdragon® 8 Gen 3 Mobile	0.894 ms	0 - 60 MB	NPU
EfficientNet-B0	ONNX	float	Snapdragon® 8 Gen 1 Mobile	2.882 ms	1 - 66 MB	NPU
EfficientNet-B0	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.259 ms	0 - 20 MB	NPU
EfficientNet-B0	ONNX	float	Qualcomm® QCS8450	2.882 ms	1 - 66 MB	NPU
EfficientNet-B0	ONNX	float	Snapdragon® 8 Elite Mobile	0.699 ms	0 - 41 MB	NPU
EfficientNet-B0	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.536 ms	1 - 38 MB	NPU
EfficientNet-B0	ONNX	float	Qualcomm® QCS9075	1.622 ms	1 - 46 MB	NPU
EfficientNet-B0	ONNX	float	Qualcomm® QCS8750	0.699 ms	0 - 41 MB	NPU
EfficientNet-B0	ONNX	float	Qualcomm® QCS7181	1.316 ms	149 - 149 MB	NPU
EfficientNet-B0	ONNX	w8a16	Snapdragon® X2 Elite	0.569 ms	181 - 181 MB	NPU
EfficientNet-B0	ONNX	w8a16	Snapdragon® X Elite	1.477 ms	149 - 149 MB	NPU
EfficientNet-B0	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	0.937 ms	26 - 99 MB	NPU
EfficientNet-B0	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	1.658 ms	0 - 74 MB	NPU
EfficientNet-B0	ONNX	w8a16	Qualcomm® QCS6490	3.517 ms	0 - 46 MB	NPU
EfficientNet-B0	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	1.436 ms	0 - 5 MB	NPU
EfficientNet-B0	ONNX	w8a16	Qualcomm® QCS8450	1.658 ms	0 - 74 MB	NPU
EfficientNet-B0	ONNX	w8a16	Qualcomm® QCS9075	1.612 ms	0 - 52 MB	NPU
EfficientNet-B0	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.548 ms	0 - 59 MB	NPU
EfficientNet-B0	ONNX	w8a16	Snapdragon® 8 Elite Mobile	0.671 ms	0 - 64 MB	NPU
EfficientNet-B0	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	1.45 ms	0 - 55 MB	NPU
EfficientNet-B0	ONNX	w8a16	Qualcomm® QCM6690	5.687 ms	0 - 62 MB	NPU
EfficientNet-B0	ONNX	w8a16	Qualcomm® QCS7790	1.45 ms	0 - 55 MB	NPU
EfficientNet-B0	ONNX	w8a16	Qualcomm® QCS8750	0.671 ms	0 - 64 MB	NPU
EfficientNet-B0	ONNX	w8a16	Qualcomm® QCS7181	1.477 ms	149 - 149 MB	NPU
EfficientNet-B0	QNN_DLC	float	Snapdragon® X2 Elite	0.846 ms	1 - 1 MB	NPU
EfficientNet-B0	QNN_DLC	float	Snapdragon® X Elite	1.763 ms	1 - 1 MB	NPU
EfficientNet-B0	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.075 ms	0 - 61 MB	NPU
EfficientNet-B0	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	3.606 ms	0 - 73 MB	NPU
EfficientNet-B0	QNN_DLC	float	Qualcomm® QCS8275	4.926 ms	1 - 36 MB	NPU
EfficientNet-B0	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	1.561 ms	0 - 4 MB	NPU
EfficientNet-B0	QNN_DLC	float	Qualcomm® QCS8450	3.606 ms	0 - 73 MB	NPU
EfficientNet-B0	QNN_DLC	float	Snapdragon® 8 Elite Mobile	0.807 ms	0 - 36 MB	NPU
EfficientNet-B0	QNN_DLC	float	Qualcomm® SA8295P	3.613 ms	0 - 43 MB	NPU
EfficientNet-B0	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.599 ms	1 - 39 MB	NPU
EfficientNet-B0	QNN_DLC	float	Qualcomm® SA7255P	4.926 ms	1 - 36 MB	NPU
EfficientNet-B0	QNN_DLC	float	Qualcomm® QCS9075	1.885 ms	1 - 3 MB	NPU
EfficientNet-B0	QNN_DLC	float	Qualcomm® QCS8750	0.807 ms	0 - 36 MB	NPU
EfficientNet-B0	QNN_DLC	float	Qualcomm® QCS7181	1.763 ms	1 - 1 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Snapdragon® X2 Elite	0.813 ms	0 - 0 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Snapdragon® X Elite	1.875 ms	0 - 0 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	1.115 ms	0 - 67 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Snapdragon® 8 Gen 1 Mobile	1.951 ms	0 - 69 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Qualcomm® QCS6490	4.128 ms	0 - 2 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Qualcomm® QCS8275	3.271 ms	0 - 47 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	1.671 ms	0 - 3 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Qualcomm® QCS8450	1.951 ms	0 - 69 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Qualcomm® QCS9075	1.855 ms	0 - 2 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.644 ms	0 - 47 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	0.774 ms	0 - 48 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Qualcomm® SA8295P	2.368 ms	0 - 46 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	1.694 ms	0 - 48 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Qualcomm® QCM6690	6.525 ms	0 - 165 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Qualcomm® SA7255P	3.271 ms	0 - 47 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Qualcomm® QCS7790	1.694 ms	0 - 48 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Qualcomm® QCS8750	0.774 ms	0 - 48 MB	NPU
EfficientNet-B0	QNN_DLC	w8a16	Qualcomm® QCS7181	1.875 ms	0 - 0 MB	NPU
EfficientNet-B0	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.069 ms	0 - 72 MB	NPU
EfficientNet-B0	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	3.621 ms	0 - 81 MB	NPU
EfficientNet-B0	TFLITE	float	Qualcomm® QCS8275	4.931 ms	0 - 41 MB	NPU
EfficientNet-B0	TFLITE	float	Qualcomm® QCS8550 (Proxy)	1.535 ms	0 - 2 MB	NPU
EfficientNet-B0	TFLITE	float	Qualcomm® SA8775P	5.739 ms	0 - 28 MB	GPU
EfficientNet-B0	TFLITE	float	Qualcomm® SA8650P	5.739 ms	0 - 28 MB	GPU
EfficientNet-B0	TFLITE	float	Qualcomm® SA8255P	5.739 ms	0 - 28 MB	GPU
EfficientNet-B0	TFLITE	float	Qualcomm® QCS8450	3.621 ms	0 - 81 MB	NPU
EfficientNet-B0	TFLITE	float	Snapdragon® 8 Elite Mobile	0.812 ms	0 - 40 MB	NPU
EfficientNet-B0	TFLITE	float	Qualcomm® SA8295P	3.632 ms	0 - 48 MB	NPU
EfficientNet-B0	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.6 ms	0 - 41 MB	NPU
EfficientNet-B0	TFLITE	float	Qualcomm® SA7255P	4.931 ms	0 - 41 MB	NPU
EfficientNet-B0	TFLITE	float	Qualcomm® QCS9075	1.888 ms	0 - 16 MB	NPU
EfficientNet-B0	TFLITE	float	Qualcomm® QCS8750	0.812 ms	0 - 40 MB	NPU

License

The license for the original implementation of EfficientNet-B0 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/EfficientNet-B0

EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

Paper • 1905.11946 • Published May 28, 2019 • 4