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	7.037 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® X Elite	14.539 ms	131 - 131 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Gen 3 Mobile	10.025 ms	0 - 478 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Gen 1 Mobile	57.144 ms	0 - 362 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	18.991 ms	2 - 7 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	14.108 ms	0 - 191 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® QCS8450	57.144 ms	0 - 362 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	19.322 ms	1 - 6 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	14.539 ms	131 - 131 MB	NPU
EfficientViT-l2-cls	ONNX	float	Qualcomm® Dragonwing™ Q-8750	7.687 ms	2 - 352 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Elite Mobile	7.687 ms	2 - 352 MB	NPU
EfficientViT-l2-cls	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	6.012 ms	2 - 377 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® X2 Elite	5.686 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® X Elite	12.824 ms	67 - 67 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	7.822 ms	0 - 323 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	14.997 ms	1 - 325 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ QCS6490	52.172 ms	0 - 4 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	11.063 ms	1 - 5 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	11.999 ms	0 - 431 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® QCS8450	14.997 ms	1 - 325 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	12.9 ms	0 - 4 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	12.824 ms	67 - 67 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	111.538 ms	0 - 459 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	15.03 ms	1 - 465 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	5.906 ms	0 - 259 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Elite Mobile	5.906 ms	0 - 259 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	4.762 ms	0 - 261 MB	NPU
EfficientViT-l2-cls	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	15.03 ms	1 - 465 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® X2 Elite	8.808 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® X Elite	18.237 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	12.296 ms	0 - 526 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	43.169 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	59.685 ms	2 - 261 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	16.891 ms	2 - 4 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8775P	21.72 ms	2 - 263 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8650P	21.72 ms	2 - 263 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8255P	21.72 ms	2 - 263 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® QCS8450	43.169 ms	0 - 355 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	21.914 ms	2 - 5 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	18.237 ms	2 - 2 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	8.935 ms	2 - 260 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA7255P	59.685 ms	2 - 261 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Qualcomm® SA8295P	34.949 ms	2 - 250 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Elite Mobile	8.935 ms	2 - 260 MB	NPU
EfficientViT-l2-cls	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	7.009 ms	2 - 270 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® X2 Elite	6.866 ms	1 - 1 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® X Elite	15.658 ms	1 - 1 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	9.35 ms	0 - 308 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	13.344 ms	1 - 4 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	27.635 ms	1 - 253 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	14.468 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	15.394 ms	3 - 5 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	15.658 ms	1 - 1 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-6690	133.923 ms	1 - 459 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-7790	18.114 ms	1 - 391 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	6.967 ms	1 - 253 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Qualcomm® SA7255P	27.635 ms	1 - 253 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	6.967 ms	1 - 253 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	5.838 ms	1 - 253 MB	NPU
EfficientViT-l2-cls	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	18.114 ms	1 - 391 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	11.661 ms	0 - 605 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	42.663 ms	0 - 441 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	21.446 ms	0 - 136 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	58.422 ms	0 - 469 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	16.166 ms	0 - 3 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8775P	20.925 ms	0 - 333 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8650P	20.925 ms	0 - 333 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8255P	20.925 ms	0 - 333 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® QCS8450	42.663 ms	0 - 441 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	20.852 ms	0 - 135 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	8.685 ms	7 - 343 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA7255P	58.422 ms	0 - 469 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Qualcomm® SA8295P	34.356 ms	1 - 322 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Elite Mobile	8.685 ms	7 - 343 MB	NPU
EfficientViT-l2-cls	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	6.846 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