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 2.50, ONNX Runtime 1.30.0	Download
ONNX	w8a16	Universal	QAIRT 2.50, ONNX Runtime 1.30.0	Download
QNN_DLC	float	Universal	QAIRT 2.50	Download
QNN_DLC	w8a16	Universal	QAIRT 2.50	Download
TFLITE	float	Universal	QAIRT 2.50	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:

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

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
EfficientFormer	ONNX	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.578 ms	0 - 44 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	0.644 ms	0 - 43 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® X2 Elite	0.564 ms	2 - 2 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® X Elite	1.196 ms	25 - 25 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® 8 Gen 3 Mobile	0.831 ms	0 - 81 MB	NPU
EfficientFormer	ONNX	float	Snapdragon® 8 Gen 1 Mobile	5.002 ms	1 - 81 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	1.606 ms	1 - 5 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.168 ms	0 - 29 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® QCS8450	5.002 ms	1 - 81 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	1.598 ms	1 - 4 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	1.196 ms	25 - 25 MB	NPU
EfficientFormer	ONNX	float	Qualcomm® Dragonwing™ Q-8750	0.644 ms	0 - 43 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.459 ms	0 - 63 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.54 ms	0 - 62 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® X2 Elite	0.483 ms	1 - 1 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® X Elite	1.261 ms	13 - 13 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	0.825 ms	0 - 87 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	1.543 ms	0 - 88 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® Dragonwing™ QCS6490	2.982 ms	0 - 4 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	1.235 ms	0 - 4 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.218 ms	0 - 16 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® QCS8450	1.543 ms	0 - 88 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	1.398 ms	0 - 3 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	1.261 ms	13 - 13 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	5.906 ms	0 - 71 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	1.342 ms	0 - 68 MB	NPU
EfficientFormer	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	0.54 ms	0 - 62 MB	NPU
EfficientFormer	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	1.342 ms	0 - 68 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.665 ms	1 - 46 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	0.763 ms	0 - 41 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® X2 Elite	0.901 ms	1 - 1 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® X Elite	1.702 ms	1 - 1 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	1.032 ms	0 - 76 MB	NPU
EfficientFormer	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	5.572 ms	0 - 81 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	1.812 ms	1 - 4 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.493 ms	1 - 2 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® SA8775P	2.045 ms	1 - 46 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® SA8650P	2.045 ms	1 - 46 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® SA8255P	2.045 ms	1 - 46 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® QCS8450	5.572 ms	0 - 81 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	1.868 ms	1 - 3 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	1.702 ms	1 - 1 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	0.763 ms	0 - 41 MB	NPU
EfficientFormer	QNN_DLC	float	Qualcomm® SA8295P	3.9 ms	0 - 42 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.536 ms	0 - 56 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	0.645 ms	0 - 50 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® X2 Elite	0.717 ms	0 - 0 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® X Elite	1.68 ms	0 - 0 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	1.022 ms	0 - 75 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	1.438 ms	0 - 4 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.469 ms	0 - 2 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® SA8775P	1.762 ms	0 - 62 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® SA8650P	1.762 ms	0 - 62 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® SA8255P	1.762 ms	0 - 62 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	1.613 ms	0 - 3 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	1.68 ms	0 - 0 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-6690	6.795 ms	0 - 173 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-7790	1.585 ms	0 - 62 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	0.645 ms	0 - 50 MB	NPU
EfficientFormer	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	1.585 ms	0 - 62 MB	NPU
EfficientFormer	TFLITE	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	0.662 ms	0 - 54 MB	NPU
EfficientFormer	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	0.764 ms	0 - 52 MB	NPU
EfficientFormer	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.032 ms	0 - 92 MB	NPU
EfficientFormer	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	5.519 ms	0 - 98 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	1.825 ms	0 - 28 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	1.466 ms	0 - 3 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® SA8775P	2.07 ms	0 - 57 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® SA8650P	2.07 ms	0 - 57 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® SA8255P	2.07 ms	0 - 57 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® QCS8450	5.519 ms	0 - 98 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	1.86 ms	0 - 27 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	0.764 ms	0 - 52 MB	NPU
EfficientFormer	TFLITE	float	Qualcomm® SA8295P	3.86 ms	0 - 51 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.