Swin-Base: Optimized for Qualcomm Devices

SwinBase 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 Swin-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
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 Swin-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 Swin-Base 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): 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
Swin-Base	ONNX	float	Snapdragon® X2 Elite	8.372 ms	2 - 2 MB	NPU
Swin-Base	ONNX	float	Snapdragon® X Elite	19.959 ms	175 - 175 MB	NPU
Swin-Base	ONNX	float	Snapdragon® 8 Gen 3 Mobile	13.114 ms	0 - 548 MB	NPU
Swin-Base	ONNX	float	Snapdragon® 8 Gen 1 Mobile	29.299 ms	1 - 1066 MB	NPU
Swin-Base	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	21.296 ms	0 - 4 MB	NPU
Swin-Base	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	19.268 ms	0 - 194 MB	NPU
Swin-Base	ONNX	float	Qualcomm® QCS8450	29.299 ms	1 - 1066 MB	NPU
Swin-Base	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	22.641 ms	1 - 4 MB	NPU
Swin-Base	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	19.959 ms	175 - 175 MB	NPU
Swin-Base	ONNX	float	Qualcomm® Dragonwing™ Q-8750	9.914 ms	0 - 400 MB	NPU
Swin-Base	ONNX	float	Snapdragon® 8 Elite Mobile	9.914 ms	0 - 400 MB	NPU
Swin-Base	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	7.974 ms	0 - 420 MB	NPU
Swin-Base	ONNX	w8a16	Snapdragon® X2 Elite	6.988 ms	1 - 1 MB	NPU
Swin-Base	ONNX	w8a16	Snapdragon® X Elite	16.28 ms	92 - 92 MB	NPU
Swin-Base	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	10.428 ms	0 - 535 MB	NPU
Swin-Base	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	20.604 ms	0 - 535 MB	NPU
Swin-Base	ONNX	w8a16	Qualcomm® Dragonwing™ QCS6490	57.993 ms	0 - 3 MB	NPU
Swin-Base	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	14.414 ms	0 - 4 MB	NPU
Swin-Base	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	15.388 ms	0 - 99 MB	NPU
Swin-Base	ONNX	w8a16	Qualcomm® QCS8450	20.604 ms	0 - 535 MB	NPU
Swin-Base	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	16.144 ms	0 - 3 MB	NPU
Swin-Base	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	16.28 ms	92 - 92 MB	NPU
Swin-Base	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	114.631 ms	0 - 689 MB	NPU
Swin-Base	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	17.805 ms	0 - 621 MB	NPU
Swin-Base	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	8.236 ms	0 - 418 MB	NPU
Swin-Base	ONNX	w8a16	Snapdragon® 8 Elite Mobile	8.236 ms	0 - 418 MB	NPU
Swin-Base	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	6.541 ms	0 - 432 MB	NPU
Swin-Base	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	17.805 ms	0 - 621 MB	NPU
Swin-Base	QNN_DLC	float	Snapdragon® X2 Elite	8.501 ms	1 - 1 MB	NPU
Swin-Base	QNN_DLC	float	Snapdragon® X Elite	19.314 ms	1 - 1 MB	NPU
Swin-Base	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	12.535 ms	0 - 516 MB	NPU
Swin-Base	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	29.092 ms	0 - 503 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	20.321 ms	1 - 4 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	53.584 ms	1 - 363 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	18.415 ms	1 - 3 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® SA8775P	21.397 ms	1 - 363 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® SA8650P	21.397 ms	1 - 363 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® SA8255P	21.397 ms	1 - 363 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® QCS8450	29.092 ms	0 - 503 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	22.02 ms	1 - 3 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	19.314 ms	1 - 1 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	9.393 ms	1 - 365 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® SA7255P	53.584 ms	1 - 363 MB	NPU
Swin-Base	QNN_DLC	float	Qualcomm® SA8295P	27.607 ms	1 - 353 MB	NPU
Swin-Base	QNN_DLC	float	Snapdragon® 8 Elite Mobile	9.393 ms	1 - 365 MB	NPU
Swin-Base	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	7.531 ms	0 - 385 MB	NPU
Swin-Base	QNN_DLC	w8a16	Snapdragon® X2 Elite	9.186 ms	0 - 0 MB	NPU
Swin-Base	QNN_DLC	w8a16	Snapdragon® X Elite	20.471 ms	0 - 0 MB	NPU
Swin-Base	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	12.755 ms	0 - 509 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	17.008 ms	0 - 3 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	35.403 ms	0 - 413 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	19.102 ms	0 - 313 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® SA8775P	19.566 ms	0 - 412 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® SA8650P	19.566 ms	0 - 412 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® SA8255P	19.566 ms	0 - 412 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	20.284 ms	0 - 2 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	20.471 ms	0 - 0 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-6690	123.34 ms	0 - 933 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-7790	21.572 ms	0 - 605 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	9.531 ms	0 - 408 MB	NPU
Swin-Base	QNN_DLC	w8a16	Qualcomm® SA7255P	35.403 ms	0 - 413 MB	NPU
Swin-Base	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	9.531 ms	0 - 408 MB	NPU
Swin-Base	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	7.609 ms	0 - 422 MB	NPU
Swin-Base	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	21.572 ms	0 - 605 MB	NPU
Swin-Base	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	12.8 ms	0 - 1057 MB	NPU
Swin-Base	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	20.905 ms	0 - 179 MB	NPU
Swin-Base	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	53.467 ms	0 - 710 MB	NPU
Swin-Base	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	19.092 ms	0 - 4 MB	NPU
Swin-Base	TFLITE	float	Qualcomm® SA8775P	21.943 ms	0 - 375 MB	NPU
Swin-Base	TFLITE	float	Qualcomm® SA8650P	21.943 ms	0 - 375 MB	NPU
Swin-Base	TFLITE	float	Qualcomm® SA8255P	21.943 ms	0 - 375 MB	NPU
Swin-Base	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	22.066 ms	0 - 178 MB	NPU
Swin-Base	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	9.65 ms	0 - 380 MB	NPU
Swin-Base	TFLITE	float	Qualcomm® SA7255P	53.467 ms	0 - 710 MB	NPU
Swin-Base	TFLITE	float	Qualcomm® SA8295P	28.376 ms	0 - 371 MB	NPU
Swin-Base	TFLITE	float	Snapdragon® 8 Elite Mobile	9.65 ms	0 - 380 MB	NPU
Swin-Base	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	7.946 ms	0 - 399 MB	NPU

License

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

References

• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
• 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.