EfficientFormer: Optimized for Qualcomm Devices

EfficientFormer is a vision transformer model that can classify images from the Imagenet dataset.

This is based on the implementation of EfficientFormer 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	w8a16	Universal	QAIRT 2.42, 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 EfficientFormer 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 EfficientFormer on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.image_classification

Model Stats:

• Model checkpoint: efficientformer_l1_300d
• Input resolution: 224x224
• Number of parameters: 12.3M
• Model size (float): 46.9 MB
• Model size (w8a16): 12.2 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
EfficientFormer	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.594 ms	0 - 50 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® X2 Elite	0.638 ms	212 - 212 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® X Elite	1.361 ms	149 - 149 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® 8 Gen 3 Mobile	0.947 ms	0 - 87 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.379 ms	0 - 34 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	0.727 ms	0 - 53 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® QCS9075	1.876 ms	1 - 46 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® QCS8750	0.727 ms	0 - 53 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® QCS7181	1.361 ms	149 - 149 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.533 ms	0 - 69 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® X2 Elite	0.562 ms	211 - 211 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® X Elite	1.478 ms	149 - 149 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	0.962 ms	0 - 94 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCS6490	139.05 ms	20 - 26 MB	CPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	1.474 ms	0 - 78 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCM6690	65.545 ms	11 - 20 MB	CPU
EfficientFormer	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	62.057 ms	21 - 30 MB	CPU
EfficientFormer	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.646 ms	0 - 60 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCS9075	1.643 ms	0 - 45 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCS7790	62.057 ms	21 - 30 MB	CPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCS8750	0.646 ms	0 - 60 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCS7181	1.478 ms	149 - 149 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.645 ms	1 - 44 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® X2 Elite	0.864 ms	1 - 1 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® X Elite	1.71 ms	1 - 1 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.04 ms	0 - 76 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® QCS8275	4.865 ms	1 - 41 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	1.471 ms	0 - 28 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® SA8775P	2.052 ms	1 - 43 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® SA8650P	2.052 ms	1 - 43 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® SA8255P	2.052 ms	1 - 43 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	5.581 ms	0 - 79 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® SA7255P	4.865 ms	1 - 41 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® SA8295P	4.001 ms	0 - 38 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	0.772 ms	0 - 41 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® QCS9075	1.987 ms	3 - 5 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® QCS8750	0.772 ms	0 - 41 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® QCS7181	1.71 ms	1 - 1 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.613 ms	0 - 66 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® X2 Elite	0.861 ms	0 - 0 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® X Elite	1.832 ms	0 - 0 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	1.082 ms	0 - 80 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® QCS8275	3.241 ms	0 - 58 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	1.545 ms	0 - 2 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® SA8775P	1.892 ms	0 - 59 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® SA8650P	1.892 ms	0 - 59 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® SA8255P	1.892 ms	0 - 59 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® QCM6690	7.02 ms	0 - 178 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® SA7255P	3.241 ms	0 - 58 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	1.694 ms	0 - 63 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.713 ms	0 - 56 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® QCS9075	1.757 ms	2 - 4 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® QCS7790	1.694 ms	0 - 63 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® QCS8750	0.713 ms	0 - 56 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® QCS7181	1.832 ms	0 - 0 MB	NPU
EfficientFormer	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.642 ms	0 - 55 MB	NPU
EfficientFormer	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.031 ms	0 - 95 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® QCS8275	4.876 ms	0 - 52 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® QCS8550 (Proxy)	1.502 ms	0 - 3 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® SA8775P	2.065 ms	0 - 54 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® SA8650P	2.065 ms	0 - 54 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® SA8255P	2.065 ms	0 - 54 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® QCS8450 (Proxy)	5.546 ms	0 - 90 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® SA7255P	4.876 ms	0 - 52 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® SA8295P	3.976 ms	0 - 47 MB	NPU
EfficientFormer	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	0.779 ms	0 - 55 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® QCS9075	1.949 ms	0 - 27 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® QCS8750	0.779 ms	0 - 55 MB	NPU

License

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

References

• Rethinking Vision Transformers for MobileNet Size and Speed
• 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.