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 1.24.1	Download
ONNX	w8a16	Universal	QAIRT 2.42, ONNX Runtime 1.24.1	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
QNN_DLC	w8a16	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.17.0	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	0.609 ms	1 - 53 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® X2 Elite	0.646 ms	25 - 25 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® X Elite	1.547 ms	24 - 24 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® 8 Gen 3 Mobile	0.945 ms	0 - 89 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® QCS8550 (Proxy)	1.322 ms	0 - 3 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® QCS9075	1.905 ms	1 - 3 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	0.705 ms	0 - 46 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.525 ms	0 - 67 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® X2 Elite	0.554 ms	12 - 12 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® X Elite	1.66 ms	12 - 12 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	0.959 ms	0 - 92 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCS6490	138.705 ms	20 - 26 MB	CPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	1.402 ms	0 - 17 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCS9075	1.619 ms	0 - 3 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCM6690	65.758 ms	23 - 31 MB	CPU
EfficientFormer	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.641 ms	0 - 56 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	62.964 ms	22 - 31 MB	CPU
EfficientFormer	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.652 ms	1 - 49 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® X2 Elite	0.913 ms	1 - 1 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® X Elite	1.767 ms	1 - 1 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.062 ms	0 - 84 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	4.914 ms	1 - 46 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	1.528 ms	1 - 3 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® QCS9075	1.973 ms	3 - 5 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	5.597 ms	0 - 81 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	0.781 ms	0 - 45 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	0.582 ms	0 - 58 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® X2 Elite	0.843 ms	0 - 0 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® X Elite	1.854 ms	0 - 0 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	1.088 ms	0 - 78 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	3.296 ms	0 - 56 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	1.616 ms	0 - 2 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® QCS9075	1.773 ms	2 - 4 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® QCM6690	6.968 ms	0 - 176 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.729 ms	0 - 55 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	1.739 ms	0 - 60 MB	NPU
EfficientFormer	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.652 ms	0 - 69 MB	NPU
EfficientFormer	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.048 ms	0 - 106 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® QCS8275 (Proxy)	4.89 ms	0 - 65 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® QCS8550 (Proxy)	1.507 ms	0 - 2 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® QCS9075	1.969 ms	0 - 27 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® QCS8450 (Proxy)	5.52 ms	0 - 101 MB	NPU
EfficientFormer	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	0.776 ms	0 - 63 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.