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	7.051 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® X Elite	14.487 ms	131 - 131 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Gen 3 Mobile	9.999 ms	0 - 477 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Gen 1 Mobile	57.286 ms	2 - 364 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	18.735 ms	2 - 7 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	14.102 ms	0 - 183 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® QCS8450	57.286 ms	2 - 364 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	18.603 ms	1 - 6 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	14.487 ms	131 - 131 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ Q-8750	7.677 ms	2 - 353 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Elite Mobile	7.677 ms	2 - 353 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	6.007 ms	1 - 375 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® X2 Elite	5.665 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® X Elite	12.849 ms	67 - 67 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	7.861 ms	1 - 325 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	15.371 ms	0 - 321 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ QCS6490	52.08 ms	0 - 4 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	11.051 ms	0 - 5 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	12.0 ms	0 - 82 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® QCS8450	15.371 ms	0 - 321 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	12.87 ms	0 - 4 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	12.849 ms	67 - 67 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	115.458 ms	0 - 455 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	14.858 ms	1 - 464 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	5.86 ms	1 - 260 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Elite Mobile	5.86 ms	1 - 260 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	4.756 ms	0 - 261 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	14.858 ms	1 - 464 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® X2 Elite	8.722 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® X Elite	18.135 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	12.145 ms	0 - 522 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	43.708 ms	0 - 355 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	22.073 ms	2 - 6 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	17.196 ms	2 - 3 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8775P	21.704 ms	2 - 262 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8650P	21.704 ms	2 - 262 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8255P	21.704 ms	2 - 262 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® QCS8450	43.708 ms	0 - 355 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	22.326 ms	2 - 5 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	18.135 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	8.927 ms	2 - 261 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA7255P	59.676 ms	2 - 261 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8295P	34.856 ms	2 - 251 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Elite Mobile	8.927 ms	2 - 261 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	7.002 ms	2 - 273 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® X2 Elite	6.88 ms	1 - 1 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® X Elite	15.673 ms	1 - 1 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	9.411 ms	0 - 309 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	13.334 ms	1 - 4 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	14.473 ms	1 - 279 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® SA8775P	14.996 ms	1 - 255 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® SA8650P	14.996 ms	1 - 255 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® SA8255P	14.996 ms	1 - 255 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	15.164 ms	3 - 5 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	15.673 ms	1 - 1 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-6690	134.292 ms	1 - 456 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-7790	17.861 ms	1 - 392 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	6.936 ms	1 - 253 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® SA7255P	27.667 ms	1 - 254 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	6.936 ms	1 - 253 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	5.836 ms	1 - 255 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	17.861 ms	1 - 392 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	11.703 ms	0 - 608 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	42.184 ms	0 - 446 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	21.732 ms	0 - 136 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	15.926 ms	0 - 3 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8775P	20.928 ms	0 - 333 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8650P	20.928 ms	0 - 333 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8255P	20.928 ms	0 - 333 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® QCS8450	42.184 ms	0 - 446 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	21.171 ms	0 - 135 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	8.73 ms	26 - 360 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA7255P	58.422 ms	0 - 469 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8295P	34.384 ms	1 - 321 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Elite Mobile	8.73 ms	26 - 360 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	6.854 ms	0 - 466 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