EfficientNet-Lite4: Optimized for Qualcomm Devices

EfficientNet-Lite4 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-Lite4 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 1.26.0	Download
ONNX	w8a8	Universal	QAIRT 2.45, ONNX Runtime 1.26.0	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 EfficientNet-Lite4 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-Lite4 on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.image_classification

Model Stats:

Model checkpoint: Imagenet
Input resolution: 300x300
Number of parameters: 13.01M
Model size (float): 51.9 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
EfficientNet-Lite4	ONNX	float	Snapdragon® X2 Elite	1.454 ms	1 - 1 MB	NPU
EfficientNet-Lite4	ONNX	float	Snapdragon® X Elite	2.739 ms	29 - 29 MB	NPU
EfficientNet-Lite4	ONNX	float	Snapdragon® 8 Gen 3 Mobile	1.905 ms	0 - 101 MB	NPU
EfficientNet-Lite4	ONNX	float	Snapdragon® 8 Gen 1 Mobile	5.519 ms	1 - 103 MB	NPU
EfficientNet-Lite4	ONNX	float	Qualcomm® QCS8550 (Proxy)	2.627 ms	0 - 54 MB	NPU
EfficientNet-Lite4	ONNX	float	Qualcomm® QCS8450	5.519 ms	1 - 103 MB	NPU
EfficientNet-Lite4	ONNX	float	Snapdragon® 8 Elite Mobile	1.5 ms	0 - 54 MB	NPU
EfficientNet-Lite4	ONNX	float	Qualcomm® QCS9075	4.109 ms	1 - 4 MB	NPU
EfficientNet-Lite4	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	1.141 ms	0 - 57 MB	NPU
EfficientNet-Lite4	ONNX	float	Qualcomm® QCS8750	1.5 ms	0 - 54 MB	NPU
EfficientNet-Lite4	ONNX	float	Qualcomm® QCS7181	2.739 ms	29 - 29 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Snapdragon® X2 Elite	0.638 ms	1 - 1 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Snapdragon® X Elite	1.259 ms	15 - 15 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.858 ms	0 - 109 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Snapdragon® 8 Gen 1 Mobile	2.079 ms	0 - 116 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Qualcomm® QCS6490	6.025 ms	0 - 3 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	1.217 ms	0 - 142 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Qualcomm® QCS8450	2.079 ms	0 - 116 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.586 ms	0 - 72 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Qualcomm® QCS9075	1.4 ms	0 - 3 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	1.597 ms	0 - 72 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Snapdragon® 8 Elite Mobile	0.685 ms	0 - 72 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Qualcomm® QCM6690	10.87 ms	0 - 193 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Qualcomm® QCS7790	1.597 ms	0 - 72 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Qualcomm® QCS8750	0.685 ms	0 - 72 MB	NPU
EfficientNet-Lite4	ONNX	w8a8	Qualcomm® QCS7181	1.259 ms	15 - 15 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Snapdragon® X2 Elite	1.923 ms	1 - 1 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Snapdragon® X Elite	3.494 ms	1 - 1 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	2.189 ms	0 - 99 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	6.871 ms	0 - 100 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Qualcomm® QCS8275	15.29 ms	1 - 53 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	3.157 ms	1 - 3 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Qualcomm® SA8775P	4.773 ms	1 - 53 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Qualcomm® SA8650P	4.773 ms	1 - 53 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Qualcomm® SA8255P	4.773 ms	1 - 53 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Qualcomm® QCS8450	6.871 ms	0 - 100 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Snapdragon® 8 Elite Mobile	1.686 ms	0 - 54 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Qualcomm® QCS9075	4.504 ms	1 - 4 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Qualcomm® SA8295P	6.451 ms	1 - 50 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.268 ms	1 - 57 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Qualcomm® SA7255P	15.29 ms	1 - 53 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Qualcomm® QCS8750	1.686 ms	0 - 54 MB	NPU
EfficientNet-Lite4	QNN_DLC	float	Qualcomm® QCS7181	3.494 ms	1 - 1 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.749 ms	0 - 0 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Snapdragon® X Elite	1.457 ms	0 - 0 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.897 ms	0 - 97 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Snapdragon® 8 Gen 1 Mobile	2.287 ms	0 - 105 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® QCS6490	7.119 ms	0 - 2 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® QCS8275	3.346 ms	0 - 62 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	1.258 ms	0 - 10 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® SA8775P	1.605 ms	0 - 62 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® SA8650P	1.605 ms	0 - 62 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® SA8255P	1.605 ms	0 - 62 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® QCS8450	2.287 ms	0 - 105 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.509 ms	0 - 66 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® SA7255P	3.346 ms	0 - 62 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® QCS9075	1.409 ms	2 - 4 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	1.627 ms	0 - 63 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Snapdragon® 8 Elite Mobile	0.641 ms	0 - 65 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® QCM6690	11.088 ms	0 - 184 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® SA8295P	2.491 ms	0 - 59 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® QCS7790	1.627 ms	0 - 63 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® QCS8750	0.641 ms	0 - 65 MB	NPU
EfficientNet-Lite4	QNN_DLC	w8a8	Qualcomm® QCS7181	1.457 ms	0 - 0 MB	NPU
EfficientNet-Lite4	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	2.193 ms	0 - 114 MB	NPU
EfficientNet-Lite4	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	6.681 ms	0 - 112 MB	NPU
EfficientNet-Lite4	TFLITE	float	Qualcomm® QCS8275	15.316 ms	0 - 65 MB	NPU
EfficientNet-Lite4	TFLITE	float	Qualcomm® QCS8550 (Proxy)	3.107 ms	0 - 2 MB	NPU
EfficientNet-Lite4	TFLITE	float	Qualcomm® SA8775P	4.819 ms	0 - 67 MB	NPU
EfficientNet-Lite4	TFLITE	float	Qualcomm® SA8650P	4.819 ms	0 - 67 MB	NPU
EfficientNet-Lite4	TFLITE	float	Qualcomm® SA8255P	4.819 ms	0 - 67 MB	NPU
EfficientNet-Lite4	TFLITE	float	Qualcomm® QCS8450	6.681 ms	0 - 112 MB	NPU
EfficientNet-Lite4	TFLITE	float	Snapdragon® 8 Elite Mobile	1.687 ms	0 - 67 MB	NPU
EfficientNet-Lite4	TFLITE	float	Qualcomm® QCS9075	4.524 ms	0 - 32 MB	NPU
EfficientNet-Lite4	TFLITE	float	Qualcomm® SA8295P	6.45 ms	0 - 55 MB	NPU
EfficientNet-Lite4	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.262 ms	0 - 67 MB	NPU
EfficientNet-Lite4	TFLITE	float	Qualcomm® SA7255P	15.316 ms	0 - 65 MB	NPU
EfficientNet-Lite4	TFLITE	float	Qualcomm® QCS8750	1.687 ms	0 - 67 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.647 ms	0 - 97 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Snapdragon® 8 Gen 1 Mobile	1.73 ms	0 - 103 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® QCS6490	5.697 ms	0 - 30 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® QCS8275	2.744 ms	0 - 57 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.904 ms	0 - 46 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® SA8775P	1.269 ms	0 - 60 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® SA8650P	1.269 ms	0 - 60 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® SA8255P	1.269 ms	0 - 60 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® QCS8450	1.73 ms	0 - 103 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.405 ms	0 - 62 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® SA7255P	2.744 ms	0 - 57 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® QCS9075	1.051 ms	0 - 17 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	1.235 ms	0 - 58 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Snapdragon® 8 Elite Mobile	0.509 ms	0 - 65 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® QCM6690	9.942 ms	0 - 180 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® SA8295P	2.116 ms	0 - 54 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® QCS7790	1.235 ms	0 - 58 MB	NPU
EfficientNet-Lite4	TFLITE	w8a8	Qualcomm® QCS8750	0.509 ms	0 - 65 MB	NPU

License

The license for the original implementation of EfficientNet-Lite4 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-Lite4

EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

Paper • 1905.11946 • Published May 28, 2019 • 4