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:

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

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
SwinV2-Base	QNN_DLC	float	Snapdragon® X2 Elite	12.465 ms	1 - 1 MB	NPU
SwinV2-Base	QNN_DLC	float	Snapdragon® X Elite	28.983 ms	1 - 1 MB	NPU
SwinV2-Base	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	19.684 ms	0 - 545 MB	NPU
SwinV2-Base	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	41.312 ms	1 - 534 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® QCS8275	74.166 ms	1 - 395 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	28.015 ms	1 - 300 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® QCS8450	41.312 ms	1 - 534 MB	NPU
SwinV2-Base	QNN_DLC	float	Snapdragon® 8 Elite Mobile	14.731 ms	1 - 391 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® SA8295P	37.74 ms	1 - 378 MB	NPU
SwinV2-Base	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	11.56 ms	0 - 427 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® SA7255P	74.166 ms	1 - 395 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® QCS9075	36.282 ms	1 - 3 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® QCS8750	14.731 ms	1 - 391 MB	NPU
SwinV2-Base	QNN_DLC	float	Qualcomm® QCS7181	28.983 ms	1 - 1 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Snapdragon® X2 Elite	12.196 ms	0 - 0 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Snapdragon® X Elite	30.76 ms	0 - 0 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	19.664 ms	0 - 2038 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® QCS8275	52.728 ms	0 - 488 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	29.221 ms	0 - 234 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® QCS9075	33.941 ms	2 - 4 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	11.393 ms	0 - 951 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	14.685 ms	0 - 904 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® SA7255P	52.728 ms	0 - 488 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® QCS8750	14.685 ms	0 - 904 MB	NPU
SwinV2-Base	QNN_DLC	w8a16	Qualcomm® QCS7181	30.76 ms	0 - 0 MB	NPU
SwinV2-Base	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	20.144 ms	0 - 2188 MB	NPU
SwinV2-Base	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	42.813 ms	0 - 676 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® QCS8275	71.114 ms	0 - 884 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® QCS8550 (Proxy)	29.364 ms	0 - 3 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® SA8775P	267.249 ms	3 - 26 MB	CPU
SwinV2-Base	TFLITE	float	Qualcomm® SA8650P	267.249 ms	3 - 26 MB	CPU
SwinV2-Base	TFLITE	float	Qualcomm® SA8255P	267.249 ms	3 - 26 MB	CPU
SwinV2-Base	TFLITE	float	Qualcomm® QCS8450	42.813 ms	0 - 676 MB	NPU
SwinV2-Base	TFLITE	float	Snapdragon® 8 Elite Mobile	14.965 ms	0 - 902 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® SA8295P	40.603 ms	0 - 875 MB	NPU
SwinV2-Base	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	11.37 ms	0 - 939 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® SA7255P	71.114 ms	0 - 884 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® QCS9075	36.727 ms	0 - 180 MB	NPU
SwinV2-Base	TFLITE	float	Qualcomm® QCS8750	14.965 ms	0 - 902 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.