SwinV2-Base: Optimized for Qualcomm Devices

SwinV2Base 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 SwinV2-Base 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
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 SwinV2-Base 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 SwinV2-Base on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.image_classification

Model Stats:

• Input resolution: 256x256
• Model checkpoint: Imagenet
• Model size (float): 339 MB
• Model size (w8a16): 90.2 MB
• Number of parameters: 88.8M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
SwinV2-Base	QNN_DLC	float	Snapdragon® X2 Elite	12.45 ms	1 - 1 MB	NPU
SwinV2-Base	QNN_DLC	float	Snapdragon® X Elite	29.063 ms	1 - 1 MB	NPU
SwinV2-Base	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	19.733 ms	0 - 545 MB	NPU
SwinV2-Base	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	41.262 ms	0 - 532 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	30.261 ms	1 - 4 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	74.05 ms	1 - 394 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	27.759 ms	1 - 421 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® SA8775P	31.516 ms	1 - 388 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® SA8650P	31.516 ms	1 - 388 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® SA8255P	31.516 ms	1 - 388 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® QCS8450	41.262 ms	0 - 532 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	31.528 ms	1 - 3 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	29.063 ms	1 - 1 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	14.696 ms	1 - 392 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® SA7255P	74.05 ms	1 - 394 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® SA8295P	37.694 ms	1 - 378 MB	NPU
SwinV2-Base	QNN_DLC	float	Snapdragon® 8 Elite Mobile	14.696 ms	1 - 392 MB	NPU
SwinV2-Base	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	11.587 ms	0 - 426 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Snapdragon® X2 Elite	12.221 ms	0 - 0 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Snapdragon® X Elite	30.739 ms	0 - 0 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	19.713 ms	0 - 2037 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	25.765 ms	0 - 3 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	52.578 ms	0 - 910 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	29.241 ms	0 - 3 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® SA8775P	30.046 ms	0 - 869 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® SA8650P	30.046 ms	0 - 869 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® SA8255P	30.046 ms	0 - 869 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	30.025 ms	0 - 2 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	30.739 ms	0 - 0 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	14.811 ms	0 - 903 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® SA7255P	52.578 ms	0 - 910 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	14.811 ms	0 - 903 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	11.419 ms	0 - 953 MB	NPU
SwinV2-Base	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	20.167 ms	0 - 2184 MB	NPU
SwinV2-Base	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	42.707 ms	0 - 677 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	29.935 ms	0 - 181 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	71.099 ms	0 - 885 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	29.461 ms	0 - 4 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® SA8775P	32.516 ms	0 - 880 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® SA8650P	32.516 ms	0 - 880 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® SA8255P	32.516 ms	0 - 880 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® QCS8450	42.707 ms	0 - 677 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	42.453 ms	0 - 181 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	14.858 ms	0 - 901 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® SA7255P	71.099 ms	0 - 885 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® SA8295P	40.6 ms	0 - 874 MB	NPU
SwinV2-Base	TFLITE	float	Snapdragon® 8 Elite Mobile	14.858 ms	0 - 901 MB	NPU
SwinV2-Base	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	11.389 ms	0 - 940 MB	NPU

License

• The license for the original implementation of SwinV2-Base can be found here.

References

• Swin Transformer V2: Scaling Up Capacity and Resolution
• Source Model Implementation

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.