EfficientViT-l2-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-l2-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-l2-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-l2-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): 243 MB
Number of parameters: 63.7M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
EfficientViT-l2-cls	ONNX	float	Snapdragon® X2 Elite	7.053 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® X Elite	14.426 ms	131 - 131 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Gen 3 Mobile	9.99 ms	0 - 478 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Gen 1 Mobile	57.004 ms	2 - 366 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	18.823 ms	2 - 7 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	14.109 ms	0 - 161 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® QCS8450	57.004 ms	2 - 366 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	18.557 ms	1 - 6 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	14.426 ms	131 - 131 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ Q-8750	7.714 ms	2 - 351 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Elite Mobile	7.714 ms	2 - 351 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	6.011 ms	0 - 373 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® X2 Elite	5.678 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® X Elite	12.811 ms	66 - 66 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	7.804 ms	1 - 324 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	14.993 ms	1 - 323 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ QCS6490	51.956 ms	0 - 4 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	11.095 ms	0 - 5 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	11.963 ms	0 - 82 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® QCS8450	14.993 ms	1 - 323 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	16.37 ms	0 - 4 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	12.811 ms	66 - 66 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	113.889 ms	0 - 456 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	14.84 ms	1 - 465 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	5.903 ms	1 - 259 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Elite Mobile	5.903 ms	1 - 259 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	4.764 ms	0 - 262 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	14.84 ms	1 - 465 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® X2 Elite	8.762 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® X Elite	18.194 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	12.346 ms	2 - 527 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	43.069 ms	0 - 355 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	22.32 ms	2 - 6 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	59.674 ms	2 - 261 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	17.267 ms	2 - 4 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8775P	21.717 ms	2 - 263 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8650P	21.717 ms	2 - 263 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8255P	21.717 ms	2 - 263 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® QCS8450	43.069 ms	0 - 355 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	21.97 ms	4 - 7 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	18.194 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	8.998 ms	2 - 260 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA7255P	59.674 ms	2 - 261 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8295P	34.905 ms	2 - 251 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Elite Mobile	8.998 ms	2 - 260 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	7.005 ms	2 - 273 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® X2 Elite	6.891 ms	1 - 1 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® X Elite	15.662 ms	1 - 1 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	9.367 ms	1 - 311 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	13.296 ms	1 - 4 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	27.644 ms	1 - 254 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	14.454 ms	1 - 3 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® SA8775P	15.001 ms	1 - 255 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® SA8650P	15.001 ms	1 - 255 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® SA8255P	15.001 ms	1 - 255 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	15.38 ms	1 - 3 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	15.662 ms	1 - 1 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-6690	137.325 ms	1 - 460 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-7790	17.833 ms	1 - 399 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	6.941 ms	1 - 252 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® SA7255P	27.644 ms	1 - 254 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	6.941 ms	1 - 252 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	5.84 ms	1 - 254 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	17.833 ms	1 - 399 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	11.768 ms	0 - 611 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	42.235 ms	0 - 441 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	21.503 ms	0 - 136 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	58.44 ms	0 - 469 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	16.241 ms	0 - 3 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8775P	20.946 ms	0 - 333 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8650P	20.946 ms	0 - 333 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8255P	20.946 ms	0 - 333 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® QCS8450	42.235 ms	0 - 441 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	21.474 ms	0 - 135 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	8.735 ms	0 - 335 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA7255P	58.44 ms	0 - 469 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8295P	34.363 ms	0 - 321 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Elite Mobile	8.735 ms	0 - 335 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	6.856 ms	0 - 468 MB	NPU

License

The license for the original implementation of EfficientViT-l2-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-l2-cls

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

Paper • 2205.14756 • Published May 29, 2022 • 1