LaMa-Dilated: Optimized for Qualcomm Devices

LaMa-Dilated is a machine learning model that allows to erase and in-paint part of given input image.

This is based on the implementation of LaMa-Dilated 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
QNN_DLC	float	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download

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

Model Details

Model Type: Model_use_case.image_editing

Model Stats:

• Model checkpoint: Dilated CelebAHQ
• Input resolution: 512x512
• Number of parameters: 45.6M
• Model size (float): 174 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
LaMa-Dilated	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	34.707 ms	0 - 386 MB	NPU
LaMa-Dilated	ONNX	float	Snapdragon® 8 Elite Mobile	41.792 ms	7 - 425 MB	NPU
LaMa-Dilated	ONNX	float	Snapdragon® X2 Elite	36.594 ms	90 - 90 MB	NPU
LaMa-Dilated	ONNX	float	Snapdragon® X Elite	78.233 ms	89 - 89 MB	NPU
LaMa-Dilated	ONNX	float	Snapdragon® X Elite	78.233 ms	89 - 89 MB	NPU
LaMa-Dilated	ONNX	float	Snapdragon® 8 Gen 3 Mobile	55.786 ms	0 - 489 MB	NPU
LaMa-Dilated	ONNX	float	Qualcomm® QCS8550 (Proxy)	78.035 ms	0 - 96 MB	NPU
LaMa-Dilated	ONNX	float	Qualcomm® QCS9075	117.926 ms	6 - 13 MB	NPU
LaMa-Dilated	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	41.792 ms	7 - 425 MB	NPU
LaMa-Dilated	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	32.664 ms	4 - 391 MB	NPU
LaMa-Dilated	QNN_DLC	float	Snapdragon® 8 Elite Mobile	40.438 ms	1 - 365 MB	NPU
LaMa-Dilated	QNN_DLC	float	Snapdragon® X2 Elite	34.126 ms	4 - 4 MB	NPU
LaMa-Dilated	QNN_DLC	float	Snapdragon® X Elite	75.357 ms	4 - 4 MB	NPU
LaMa-Dilated	QNN_DLC	float	Snapdragon® X Elite	75.357 ms	4 - 4 MB	NPU
LaMa-Dilated	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	52.594 ms	0 - 455 MB	NPU
LaMa-Dilated	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	402.982 ms	1 - 364 MB	NPU
LaMa-Dilated	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	74.822 ms	4 - 6 MB	NPU
LaMa-Dilated	QNN_DLC	float	Qualcomm® SA8775P	106.966 ms	1 - 364 MB	NPU
LaMa-Dilated	QNN_DLC	float	Qualcomm® SA8775P	106.966 ms	1 - 364 MB	NPU
LaMa-Dilated	QNN_DLC	float	Qualcomm® SA8775P	106.966 ms	1 - 364 MB	NPU
LaMa-Dilated	QNN_DLC	float	Qualcomm® QCS9075	118.032 ms	4 - 13 MB	NPU
LaMa-Dilated	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	134.778 ms	4 - 409 MB	NPU
LaMa-Dilated	QNN_DLC	float	Qualcomm® SA7255P	402.982 ms	1 - 364 MB	NPU
LaMa-Dilated	QNN_DLC	float	Qualcomm® SA8295P	109.521 ms	1 - 299 MB	NPU
LaMa-Dilated	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	40.438 ms	1 - 365 MB	NPU
LaMa-Dilated	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	32.225 ms	1 - 409 MB	NPU
LaMa-Dilated	TFLITE	float	Snapdragon® 8 Elite Mobile	40.232 ms	2 - 389 MB	NPU
LaMa-Dilated	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	52.546 ms	2 - 537 MB	NPU
LaMa-Dilated	TFLITE	float	Qualcomm® QCS8275 (Proxy)	404.195 ms	3 - 381 MB	NPU
LaMa-Dilated	TFLITE	float	Qualcomm® QCS8550 (Proxy)	74.173 ms	3 - 133 MB	NPU
LaMa-Dilated	TFLITE	float	Qualcomm® SA8775P	106.487 ms	3 - 380 MB	NPU
LaMa-Dilated	TFLITE	float	Qualcomm® SA8775P	106.487 ms	3 - 380 MB	NPU
LaMa-Dilated	TFLITE	float	Qualcomm® SA8775P	106.487 ms	3 - 380 MB	NPU
LaMa-Dilated	TFLITE	float	Qualcomm® QCS9075	118.569 ms	2 - 104 MB	NPU
LaMa-Dilated	TFLITE	float	Qualcomm® QCS8450 (Proxy)	133.429 ms	3 - 459 MB	NPU
LaMa-Dilated	TFLITE	float	Qualcomm® SA7255P	404.195 ms	3 - 381 MB	NPU
LaMa-Dilated	TFLITE	float	Qualcomm® SA8295P	110.158 ms	3 - 311 MB	NPU
LaMa-Dilated	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	40.232 ms	2 - 389 MB	NPU

License

• The license for the original implementation of LaMa-Dilated can be found here.

References

• Resolution-robust Large Mask Inpainting with Fourier Convolutions
• 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.