XLSR: Optimized for Qualcomm Devices

XLSR is designed for lightweight real-time upscaling of images.

This is based on the implementation of XLSR 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 1.25.0	Download
ONNX	w8a8	Universal	QAIRT 2.45, ONNX Runtime 1.25.0	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
QNN_DLC	w8a8	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download
TFLITE	w8a8	Universal	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit XLSR 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 XLSR on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.super_resolution

Model Stats:

Model checkpoint: xlsr_3x_checkpoint
Input resolution: 128x128
Number of parameters: 28.0K
Model size (float): 115 KB
Model size (w8a8): 45.6 KB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
XLSR	ONNX	float	Snapdragon® X2 Elite	0.37 ms	213 - 213 MB	NPU
XLSR	ONNX	float	Snapdragon® X Elite	0.876 ms	181 - 181 MB	NPU
XLSR	ONNX	float	Snapdragon® 8 Gen 3 Mobile	0.576 ms	0 - 18 MB	NPU
XLSR	ONNX	float	Snapdragon® 8 Gen 1 Mobile	1.198 ms	0 - 26 MB	NPU
XLSR	ONNX	float	Qualcomm® QCS8550 (Proxy)	0.873 ms	0 - 63 MB	NPU
XLSR	ONNX	float	Qualcomm® QCS8450	1.198 ms	0 - 26 MB	NPU
XLSR	ONNX	float	Snapdragon® 8 Elite Mobile	0.479 ms	0 - 19 MB	NPU
XLSR	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	0.399 ms	0 - 16 MB	NPU
XLSR	ONNX	float	Qualcomm® QCS9075	1.195 ms	6 - 51 MB	NPU
XLSR	ONNX	float	Qualcomm® QCS8750	0.479 ms	0 - 19 MB	NPU
XLSR	ONNX	float	Qualcomm® QCS7181	0.876 ms	181 - 181 MB	NPU
XLSR	ONNX	w8a8	Snapdragon® X2 Elite	0.256 ms	213 - 213 MB	NPU
XLSR	ONNX	w8a8	Snapdragon® X Elite	0.636 ms	181 - 181 MB	NPU
XLSR	ONNX	w8a8	Snapdragon® 8 Gen 3 Mobile	0.412 ms	0 - 21 MB	NPU
XLSR	ONNX	w8a8	Snapdragon® 8 Gen 1 Mobile	0.727 ms	0 - 22 MB	NPU
XLSR	ONNX	w8a8	Qualcomm® QCS6490	2.371 ms	0 - 45 MB	NPU
XLSR	ONNX	w8a8	Qualcomm® QCS8550 (Proxy)	0.643 ms	0 - 16 MB	NPU
XLSR	ONNX	w8a8	Qualcomm® QCS8450	0.727 ms	0 - 22 MB	NPU
XLSR	ONNX	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.274 ms	0 - 20 MB	NPU
XLSR	ONNX	w8a8	Snapdragon® 7 Gen 4 Mobile	0.672 ms	0 - 18 MB	NPU
XLSR	ONNX	w8a8	Qualcomm® QCM6690	2.833 ms	0 - 133 MB	NPU
XLSR	ONNX	w8a8	Qualcomm® QCS9075	0.808 ms	0 - 45 MB	NPU
XLSR	ONNX	w8a8	Snapdragon® 8 Elite Mobile	0.31 ms	0 - 29 MB	NPU
XLSR	ONNX	w8a8	Qualcomm® QCS7790	0.672 ms	0 - 18 MB	NPU
XLSR	ONNX	w8a8	Qualcomm® QCS8750	0.31 ms	0 - 29 MB	NPU
XLSR	ONNX	w8a8	Qualcomm® QCS7181	0.636 ms	181 - 181 MB	NPU
XLSR	QNN_DLC	float	Snapdragon® X2 Elite	0.41 ms	0 - 0 MB	NPU
XLSR	QNN_DLC	float	Snapdragon® X Elite	0.882 ms	0 - 0 MB	NPU
XLSR	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	0.464 ms	0 - 29 MB	NPU
XLSR	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	1.075 ms	0 - 35 MB	NPU
XLSR	QNN_DLC	float	Qualcomm® QCS8275	2.211 ms	0 - 21 MB	NPU
XLSR	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	0.764 ms	0 - 61 MB	NPU
XLSR	QNN_DLC	float	Qualcomm® QCS8450	1.075 ms	0 - 35 MB	NPU
XLSR	QNN_DLC	float	Snapdragon® 8 Elite Mobile	0.362 ms	0 - 21 MB	NPU
XLSR	QNN_DLC	float	Qualcomm® SA8295P	1.421 ms	0 - 18 MB	NPU
XLSR	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	0.34 ms	0 - 23 MB	NPU
XLSR	QNN_DLC	float	Qualcomm® SA7255P	2.211 ms	0 - 21 MB	NPU
XLSR	QNN_DLC	float	Qualcomm® QCS9075	1.102 ms	0 - 5 MB	NPU
XLSR	QNN_DLC	float	Qualcomm® QCS8750	0.362 ms	0 - 21 MB	NPU
XLSR	QNN_DLC	float	Qualcomm® QCS7181	0.882 ms	0 - 0 MB	NPU
XLSR	QNN_DLC	w8a8	Snapdragon® X2 Elite	0.278 ms	0 - 0 MB	NPU
XLSR	QNN_DLC	w8a8	Snapdragon® X Elite	0.481 ms	0 - 0 MB	NPU
XLSR	QNN_DLC	w8a8	Snapdragon® 8 Gen 3 Mobile	0.251 ms	0 - 30 MB	NPU
XLSR	QNN_DLC	w8a8	Snapdragon® 8 Gen 1 Mobile	0.474 ms	0 - 35 MB	NPU
XLSR	QNN_DLC	w8a8	Qualcomm® QCS6490	1.395 ms	0 - 2 MB	NPU
XLSR	QNN_DLC	w8a8	Qualcomm® QCS8275	0.982 ms	0 - 20 MB	NPU
XLSR	QNN_DLC	w8a8	Qualcomm® QCS8550 (Proxy)	0.399 ms	0 - 35 MB	NPU
XLSR	QNN_DLC	w8a8	Qualcomm® QCS8450	0.474 ms	0 - 35 MB	NPU
XLSR	QNN_DLC	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.182 ms	0 - 21 MB	NPU
XLSR	QNN_DLC	w8a8	Snapdragon® 7 Gen 4 Mobile	0.419 ms	0 - 19 MB	NPU
XLSR	QNN_DLC	w8a8	Qualcomm® QCM6690	1.606 ms	0 - 19 MB	NPU
XLSR	QNN_DLC	w8a8	Qualcomm® QCS9075	0.558 ms	0 - 2 MB	NPU
XLSR	QNN_DLC	w8a8	Qualcomm® SA7255P	0.982 ms	0 - 20 MB	NPU
XLSR	QNN_DLC	w8a8	Snapdragon® 8 Elite Mobile	0.215 ms	0 - 26 MB	NPU
XLSR	QNN_DLC	w8a8	Qualcomm® SA8295P	0.786 ms	0 - 18 MB	NPU
XLSR	QNN_DLC	w8a8	Qualcomm® QCS7790	0.419 ms	0 - 19 MB	NPU
XLSR	QNN_DLC	w8a8	Qualcomm® QCS8750	0.215 ms	0 - 26 MB	NPU
XLSR	QNN_DLC	w8a8	Qualcomm® QCS7181	0.481 ms	0 - 0 MB	NPU
XLSR	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	1.395 ms	0 - 31 MB	NPU
XLSR	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	2.511 ms	0 - 36 MB	NPU
XLSR	TFLITE	float	Qualcomm® QCS8275	4.716 ms	3 - 25 MB	NPU
XLSR	TFLITE	float	Qualcomm® QCS8550 (Proxy)	2.197 ms	0 - 1 MB	NPU
XLSR	TFLITE	float	Qualcomm® SA8775P	3.248 ms	0 - 29 MB	GPU
XLSR	TFLITE	float	Qualcomm® SA8650P	3.248 ms	0 - 29 MB	GPU
XLSR	TFLITE	float	Qualcomm® SA8255P	3.248 ms	0 - 29 MB	GPU
XLSR	TFLITE	float	Qualcomm® QCS8450	2.511 ms	0 - 36 MB	NPU
XLSR	TFLITE	float	Snapdragon® 8 Elite Mobile	0.973 ms	0 - 22 MB	NPU
XLSR	TFLITE	float	Qualcomm® SA8295P	3.069 ms	0 - 18 MB	NPU
XLSR	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	0.794 ms	0 - 22 MB	NPU
XLSR	TFLITE	float	Qualcomm® SA7255P	4.716 ms	3 - 25 MB	NPU
XLSR	TFLITE	float	Qualcomm® QCS9075	2.58 ms	3 - 8 MB	NPU
XLSR	TFLITE	float	Qualcomm® QCS8750	0.973 ms	0 - 22 MB	NPU
XLSR	TFLITE	w8a8	Snapdragon® 8 Gen 3 Mobile	0.274 ms	0 - 29 MB	NPU
XLSR	TFLITE	w8a8	Snapdragon® 8 Gen 1 Mobile	0.47 ms	0 - 35 MB	NPU
XLSR	TFLITE	w8a8	Qualcomm® QCS6490	1.098 ms	0 - 3 MB	NPU
XLSR	TFLITE	w8a8	Qualcomm® QCS8275	0.95 ms	0 - 21 MB	NPU
XLSR	TFLITE	w8a8	Qualcomm® QCS8550 (Proxy)	0.404 ms	0 - 2 MB	NPU
XLSR	TFLITE	w8a8	Qualcomm® SA8775P	3.877 ms	4 - 33 MB	GPU
XLSR	TFLITE	w8a8	Qualcomm® SA8650P	3.877 ms	4 - 33 MB	GPU
XLSR	TFLITE	w8a8	Qualcomm® SA8255P	3.877 ms	4 - 33 MB	GPU
XLSR	TFLITE	w8a8	Qualcomm® QCS8450	0.47 ms	0 - 35 MB	NPU
XLSR	TFLITE	w8a8	Snapdragon® 8 Elite Gen 5 Mobile	0.181 ms	0 - 21 MB	NPU
XLSR	TFLITE	w8a8	Snapdragon® 7 Gen 4 Mobile	0.42 ms	0 - 20 MB	NPU
XLSR	TFLITE	w8a8	Qualcomm® QCM6690	1.621 ms	0 - 19 MB	NPU
XLSR	TFLITE	w8a8	Qualcomm® QCS9075	0.581 ms	0 - 3 MB	NPU
XLSR	TFLITE	w8a8	Qualcomm® SA7255P	0.95 ms	0 - 21 MB	NPU
XLSR	TFLITE	w8a8	Snapdragon® 8 Elite Mobile	0.228 ms	0 - 21 MB	NPU
XLSR	TFLITE	w8a8	Qualcomm® SA8295P	0.804 ms	0 - 18 MB	NPU
XLSR	TFLITE	w8a8	Qualcomm® QCS7790	0.42 ms	0 - 20 MB	NPU
XLSR	TFLITE	w8a8	Qualcomm® QCS8750	0.228 ms	0 - 21 MB	NPU

License

The license for the original implementation of XLSR 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/XLSR

Extremely Lightweight Quantization Robust Real-Time Single-Image Super Resolution for Mobile Devices

Paper • 2105.10288 • Published May 21, 2021