ESRGAN: Optimized for Qualcomm Devices

ESRGAN is a machine learning model that upscales an image with minimal loss in quality.

This is based on the implementation of ESRGAN 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 2.42, ONNX Runtime 1.24.3	Download
ONNX	w8a16	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	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 ESRGAN 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 ESRGAN on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.super_resolution

Model Stats:

• Model checkpoint: ESRGAN_x4
• Input resolution: 128x128
• Number of parameters: 16.7M
• Model size (float): 63.9 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
ESRGAN	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	27.953 ms	7 - 354 MB	NPU
ESRGAN	ONNX	float	Snapdragon® X2 Elite	34.428 ms	37 - 37 MB	NPU
ESRGAN	ONNX	float	Snapdragon® X Elite	65.458 ms	37 - 37 MB	NPU
ESRGAN	ONNX	float	Snapdragon® 8 Gen 3 Mobile	49.658 ms	6 - 791 MB	NPU
ESRGAN	ONNX	float	Qualcomm® QCS8550 (Proxy)	68.828 ms	0 - 44 MB	NPU
ESRGAN	ONNX	float	Qualcomm® QCS9075	107.687 ms	6 - 9 MB	NPU
ESRGAN	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	38.299 ms	0 - 331 MB	NPU
ESRGAN	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	18.967 ms	3 - 1057 MB	NPU
ESRGAN	ONNX	w8a16	Snapdragon® X2 Elite	22.026 ms	29 - 29 MB	NPU
ESRGAN	ONNX	w8a16	Snapdragon® X Elite	43.64 ms	26 - 26 MB	NPU
ESRGAN	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	31.84 ms	3 - 1278 MB	NPU
ESRGAN	ONNX	w8a16	Qualcomm® QCS6490	14967.555 ms	201 - 206 MB	CPU
ESRGAN	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	42.202 ms	0 - 180 MB	NPU
ESRGAN	ONNX	w8a16	Qualcomm® QCS9075	45.32 ms	3 - 6 MB	NPU
ESRGAN	ONNX	w8a16	Qualcomm® QCM6690	7966.267 ms	185 - 206 MB	CPU
ESRGAN	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	26.423 ms	0 - 904 MB	NPU
ESRGAN	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	7721.104 ms	147 - 165 MB	CPU
ESRGAN	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	25.333 ms	0 - 330 MB	NPU
ESRGAN	QNN_DLC	float	Snapdragon® X2 Elite	34.329 ms	0 - 0 MB	NPU
ESRGAN	QNN_DLC	float	Snapdragon® X Elite	64.929 ms	0 - 0 MB	NPU
ESRGAN	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	48.962 ms	0 - 703 MB	NPU
ESRGAN	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	451.986 ms	0 - 353 MB	NPU
ESRGAN	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	62.761 ms	0 - 638 MB	NPU
ESRGAN	QNN_DLC	float	Qualcomm® SA8775P	105.513 ms	0 - 353 MB	NPU
ESRGAN	QNN_DLC	float	Qualcomm® QCS9075	106.562 ms	0 - 5 MB	NPU
ESRGAN	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	116.005 ms	0 - 709 MB	NPU
ESRGAN	QNN_DLC	float	Qualcomm® SA7255P	451.986 ms	0 - 353 MB	NPU
ESRGAN	QNN_DLC	float	Qualcomm® SA8295P	111.344 ms	0 - 358 MB	NPU
ESRGAN	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	37.68 ms	0 - 334 MB	NPU
ESRGAN	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	16.166 ms	0 - 992 MB	NPU
ESRGAN	QNN_DLC	w8a16	Snapdragon® X2 Elite	21.871 ms	0 - 0 MB	NPU
ESRGAN	QNN_DLC	w8a16	Snapdragon® X Elite	43.097 ms	0 - 0 MB	NPU
ESRGAN	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	31.373 ms	0 - 1036 MB	NPU
ESRGAN	QNN_DLC	w8a16	Qualcomm® QCS6490	240.601 ms	0 - 3 MB	NPU
ESRGAN	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	132.912 ms	0 - 671 MB	NPU
ESRGAN	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	41.531 ms	0 - 1010 MB	NPU
ESRGAN	QNN_DLC	w8a16	Qualcomm® SA8775P	37.632 ms	0 - 672 MB	NPU
ESRGAN	QNN_DLC	w8a16	Qualcomm® QCS9075	44.048 ms	2 - 5 MB	NPU
ESRGAN	QNN_DLC	w8a16	Qualcomm® QCM6690	1128.818 ms	0 - 637 MB	NPU
ESRGAN	QNN_DLC	w8a16	Qualcomm® QCS8450 (Proxy)	79.535 ms	1 - 1112 MB	NPU
ESRGAN	QNN_DLC	w8a16	Qualcomm® SA7255P	132.912 ms	0 - 671 MB	NPU
ESRGAN	QNN_DLC	w8a16	Qualcomm® SA8295P	65.055 ms	0 - 716 MB	NPU
ESRGAN	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	27.907 ms	0 - 846 MB	NPU
ESRGAN	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	88.839 ms	0 - 697 MB	NPU
ESRGAN	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	26.634 ms	3 - 370 MB	NPU
ESRGAN	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	48.723 ms	3 - 746 MB	NPU
ESRGAN	TFLITE	float	Qualcomm® QCS8275 (Proxy)	452.061 ms	3 - 396 MB	NPU
ESRGAN	TFLITE	float	Qualcomm® QCS8550 (Proxy)	64.493 ms	3 - 6 MB	NPU
ESRGAN	TFLITE	float	Qualcomm® SA8775P	105.551 ms	3 - 396 MB	NPU
ESRGAN	TFLITE	float	Qualcomm® QCS9075	108.57 ms	1 - 45 MB	NPU
ESRGAN	TFLITE	float	Qualcomm® QCS8450 (Proxy)	113.984 ms	4 - 745 MB	NPU
ESRGAN	TFLITE	float	Qualcomm® SA7255P	452.061 ms	3 - 396 MB	NPU
ESRGAN	TFLITE	float	Qualcomm® SA8295P	111.391 ms	1 - 389 MB	NPU
ESRGAN	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	37.781 ms	3 - 367 MB	NPU

License

• The license for the original implementation of ESRGAN can be found here.

References

• ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks
• 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.