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 2.50, ONNX Runtime 1.27.1	Download
ONNX	w8a16	Universal	QAIRT 2.50, ONNX Runtime 1.27.1	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 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® 8 Elite Gen 5 For Galaxy Mobile	2.45 ms	0 - 74 MB	NPU
EfficientViT-b2-cls	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	2.697 ms	0 - 74 MB	NPU
EfficientViT-b2-cls	ONNX	float	Snapdragon® X2 Elite	2.517 ms	2 - 2 MB	NPU
EfficientViT-b2-cls	ONNX	float	Snapdragon® X Elite	5.063 ms	50 - 50 MB	NPU
EfficientViT-b2-cls	ONNX	float	Snapdragon® 8 Gen 3 Mobile	3.229 ms	0 - 146 MB	NPU
EfficientViT-b2-cls	ONNX	float	Snapdragon® 8 Gen 1 Mobile	5.662 ms	1 - 146 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	5.103 ms	0 - 5 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	4.887 ms	0 - 58 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® QCS8450	5.662 ms	1 - 146 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	5.594 ms	0 - 4 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	5.063 ms	50 - 50 MB	NPU
EfficientViT-b2-cls	ONNX	float	Qualcomm® Dragonwing™ Q-8750	2.697 ms	0 - 74 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	2.096 ms	0 - 121 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	2.443 ms	0 - 120 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® X2 Elite	2.233 ms	1 - 1 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® X Elite	4.683 ms	27 - 27 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	3.006 ms	0 - 164 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	5.375 ms	0 - 176 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ QCS6490	11.074 ms	0 - 4 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	4.077 ms	0 - 4 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	4.488 ms	0 - 30 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® QCS8450	5.375 ms	0 - 176 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	4.752 ms	0 - 4 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	4.683 ms	27 - 27 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	24.151 ms	0 - 239 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	4.809 ms	0 - 231 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	2.443 ms	0 - 120 MB	NPU
EfficientViT-b2-cls	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	4.809 ms	0 - 231 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	2.075 ms	1 - 72 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	2.358 ms	0 - 69 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® X2 Elite	2.454 ms	1 - 1 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® X Elite	4.869 ms	1 - 1 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	2.948 ms	0 - 144 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	5.882 ms	0 - 142 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	4.61 ms	1 - 4 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	4.512 ms	1 - 2 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® SA8650P	6.803 ms	1 - 69 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® SA8255P	6.803 ms	1 - 69 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® QCS8450	5.882 ms	0 - 142 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	5.145 ms	3 - 5 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	4.869 ms	1 - 1 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	2.358 ms	0 - 69 MB	NPU
EfficientViT-b2-cls	QNN_DLC	float	Qualcomm® SA8295P	5.934 ms	0 - 72 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	2.094 ms	0 - 112 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	2.458 ms	0 - 107 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® X2 Elite	2.529 ms	0 - 0 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® X Elite	5.486 ms	0 - 0 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	3.359 ms	0 - 151 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	4.493 ms	0 - 4 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	5.044 ms	0 - 2 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® SA8650P	5.456 ms	0 - 106 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® SA8255P	5.456 ms	0 - 106 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	5.326 ms	2 - 5 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	5.486 ms	0 - 0 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-6690	24.513 ms	0 - 224 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-7790	5.327 ms	0 - 215 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	2.458 ms	0 - 107 MB	NPU
EfficientViT-b2-cls	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	5.327 ms	0 - 215 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	2.062 ms	0 - 115 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	2.353 ms	0 - 110 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	2.957 ms	0 - 179 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	5.853 ms	0 - 181 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	4.642 ms	0 - 53 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	4.478 ms	0 - 3 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® SA8650P	6.819 ms	0 - 111 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® SA8255P	6.819 ms	0 - 111 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® QCS8450	5.853 ms	0 - 181 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	5.124 ms	0 - 52 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	2.353 ms	0 - 110 MB	NPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® SA7255P	30.281 ms	0 - 27 MB	GPU
EfficientViT-b2-cls	TFLITE	float	Qualcomm® SA8295P	5.914 ms	0 - 107 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