EfficientViT-b2-cls: Optimized for Qualcomm Devices

EfficientViT 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 EfficientViT-b2-cls 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 1.27.1	Download
ONNX	w8a16	Universal	QAIRT 2.45, ONNX Runtime 1.27.1	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 EfficientViT-b2-cls 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 EfficientViT-b2-cls on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.image_classification

Model Stats:

Input resolution: 224x224
Model checkpoint: Imagenet
Model size (float): 92.9 MB
Number of parameters: 24.3M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
EfficientViT-b2-cls	ONNX	float	Snapdragon® X2 Elite	2.526 ms	2 - 2 MB	NPU
EfficientViT-b2-cls	ONNX	float	Snapdragon® X Elite	5.057 ms	49 - 49 MB	NPU
EfficientViT-b2-cls	ONNX	float	Snapdragon® 8 Gen 3 Mobile	3.235 ms	0 - 140 MB	NPU
EfficientViT-b2-cls	ONNX	float	Snapdragon® 8 Gen 1 Mobile	6.326 ms	1 - 143 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	5.066 ms	0 - 4 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	4.894 ms	0 - 63 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® QCS8450	6.326 ms	1 - 143 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	5.527 ms	0 - 4 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	5.057 ms	49 - 49 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® Dragonwing™ Q-8750	2.695 ms	0 - 70 MB	NPU
EfficientViT-b2-cls	ONNX	float	Snapdragon® 8 Elite Mobile	2.695 ms	0 - 70 MB	NPU
EfficientViT-b2-cls	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	2.444 ms	0 - 71 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® X2 Elite	2.607 ms	1 - 1 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® X Elite	4.854 ms	27 - 27 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	3.081 ms	0 - 157 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	5.773 ms	0 - 169 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ QCS6490	15.321 ms	0 - 3 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	4.228 ms	0 - 4 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	4.617 ms	0 - 34 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® QCS8450	5.773 ms	0 - 169 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	4.911 ms	0 - 3 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	4.854 ms	27 - 27 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	22.184 ms	0 - 256 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	4.997 ms	0 - 116 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	2.522 ms	0 - 114 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® 8 Elite Mobile	2.522 ms	0 - 114 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	2.169 ms	0 - 115 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	4.997 ms	0 - 116 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® X2 Elite	2.956 ms	1 - 1 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® X Elite	6.186 ms	1 - 1 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	3.71 ms	0 - 141 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	7.141 ms	0 - 144 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	6.186 ms	1 - 4 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	12.826 ms	1 - 67 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	5.431 ms	1 - 2 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® SA8775P	6.8 ms	1 - 69 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® SA8650P	6.8 ms	1 - 69 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® SA8255P	6.8 ms	1 - 69 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® QCS8450	7.141 ms	0 - 144 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	6.537 ms	1 - 3 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	6.186 ms	1 - 1 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	2.82 ms	1 - 70 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® SA7255P	12.826 ms	1 - 67 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® SA8295P	7.419 ms	1 - 73 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® 8 Elite Mobile	2.82 ms	1 - 70 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	2.341 ms	1 - 74 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® X2 Elite	2.597 ms	0 - 0 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® X Elite	5.478 ms	0 - 0 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	3.285 ms	0 - 143 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	4.424 ms	0 - 3 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	9.237 ms	0 - 104 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	4.956 ms	0 - 2 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® SA8775P	5.459 ms	0 - 106 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® SA8650P	5.459 ms	0 - 106 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® SA8255P	5.459 ms	0 - 106 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	5.286 ms	2 - 4 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	5.478 ms	0 - 0 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-6690	21.447 ms	0 - 229 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-7790	5.18 ms	0 - 102 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	2.52 ms	0 - 107 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® SA7255P	9.237 ms	0 - 104 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	2.52 ms	0 - 107 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	2.119 ms	0 - 106 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	5.18 ms	0 - 102 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	3.728 ms	0 - 184 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	7.148 ms	0 - 186 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	6.199 ms	0 - 52 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	13.009 ms	0 - 110 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	5.451 ms	0 - 3 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® SA8775P	6.79 ms	0 - 112 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® SA8650P	6.79 ms	0 - 112 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® SA8255P	6.79 ms	0 - 112 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® QCS8450	7.148 ms	0 - 186 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	6.49 ms	0 - 52 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	2.785 ms	0 - 114 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® SA7255P	13.009 ms	0 - 110 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® SA8295P	7.495 ms	0 - 115 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Snapdragon® 8 Elite Mobile	2.785 ms	0 - 114 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	2.359 ms	0 - 118 MB	NPU

License

The license for the original implementation of EfficientViT-b2-cls 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/EfficientViT-b2-cls

EfficientViT: Lightweight Multi-Scale Attention for On-Device Semantic Segmentation

Paper • 2205.14756 • Published May 29, 2022 • 1