Real-ESRGAN-General-x4v3: Optimized for Qualcomm Devices
Real-ESRGAN is a machine learning model that upscales an image with minimal loss in quality.
This is based on the implementation of Real-ESRGAN-General-x4v3 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 1.24.1 | Download |
| ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| QNN_DLC | float | Universal | QAIRT 2.43 | Download |
| QNN_DLC | w8a8 | Universal | QAIRT 2.43 | Download |
| TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | Download |
| TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | Download |
For more device-specific assets and performance metrics, visit Real-ESRGAN-General-x4v3 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 Real-ESRGAN-General-x4v3 on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.super_resolution
Model Stats:
- Model checkpoint: realesr-general-x4v3
- Input resolution: 128x128
- Number of parameters: 1.21M
- Model size (float): 4.65 MB
- Model size (w8a8): 1.25 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| Real-ESRGAN-General-x4v3 | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.657 ms | 0 - 152 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | float | Snapdragon® X2 Elite | 6.422 ms | 4 - 4 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | float | Snapdragon® X Elite | 6.398 ms | 8 - 8 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 4.549 ms | 7 - 195 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | float | Qualcomm® QCS8550 (Proxy) | 6.051 ms | 6 - 8 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | float | Qualcomm® QCS9075 | 9.764 ms | 6 - 9 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.57 ms | 0 - 151 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.688 ms | 0 - 157 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | w8a8 | Snapdragon® X2 Elite | 0.877 ms | 0 - 0 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | w8a8 | Snapdragon® X Elite | 1.927 ms | 2 - 2 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.227 ms | 0 - 169 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | w8a8 | Qualcomm® QCS6490 | 154.404 ms | 23 - 30 MB | CPU |
| Real-ESRGAN-General-x4v3 | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.727 ms | 0 - 87 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | w8a8 | Qualcomm® QCS9075 | 1.949 ms | 0 - 3 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | w8a8 | Qualcomm® QCM6690 | 146.911 ms | 26 - 35 MB | CPU |
| Real-ESRGAN-General-x4v3 | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.974 ms | 0 - 45 MB | NPU |
| Real-ESRGAN-General-x4v3 | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 117.508 ms | 32 - 40 MB | CPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.302 ms | 0 - 158 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Snapdragon® X2 Elite | 3.508 ms | 0 - 0 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Snapdragon® X Elite | 6.36 ms | 0 - 0 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 4.279 ms | 0 - 185 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 33.177 ms | 0 - 151 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 5.851 ms | 0 - 2 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Qualcomm® SA8775P | 42.216 ms | 0 - 151 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Qualcomm® QCS9075 | 9.664 ms | 0 - 5 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 8.867 ms | 0 - 92 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Qualcomm® SA7255P | 33.177 ms | 0 - 151 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Qualcomm® SA8295P | 9.729 ms | 0 - 50 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.205 ms | 0 - 153 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.657 ms | 0 - 151 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.988 ms | 0 - 0 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Snapdragon® X Elite | 1.954 ms | 0 - 0 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.23 ms | 0 - 162 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 6.207 ms | 0 - 2 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 5.669 ms | 0 - 147 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.746 ms | 0 - 2 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Qualcomm® SA8775P | 2.066 ms | 0 - 151 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 1.975 ms | 0 - 2 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 24.902 ms | 0 - 147 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 2.891 ms | 0 - 66 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Qualcomm® SA7255P | 5.669 ms | 0 - 147 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Qualcomm® SA8295P | 3.255 ms | 0 - 44 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.92 ms | 0 - 34 MB | NPU |
| Real-ESRGAN-General-x4v3 | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 2.411 ms | 0 - 147 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.302 ms | 0 - 163 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 4.288 ms | 0 - 191 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 33.2 ms | 3 - 157 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 5.847 ms | 3 - 5 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | float | Qualcomm® SA8775P | 9.454 ms | 3 - 160 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | float | Qualcomm® QCS9075 | 9.565 ms | 1 - 9 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 8.877 ms | 3 - 95 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | float | Qualcomm® SA7255P | 33.2 ms | 3 - 157 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | float | Qualcomm® SA8295P | 9.753 ms | 3 - 57 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.16 ms | 0 - 157 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.661 ms | 0 - 152 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 1.229 ms | 0 - 164 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Qualcomm® QCS6490 | 6.295 ms | 1 - 5 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 5.741 ms | 1 - 149 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 1.768 ms | 0 - 2 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Qualcomm® SA8775P | 8.97 ms | 1 - 150 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Qualcomm® QCS9075 | 1.993 ms | 0 - 4 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Qualcomm® QCM6690 | 25.362 ms | 1 - 150 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 2.852 ms | 0 - 70 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Qualcomm® SA7255P | 5.741 ms | 1 - 149 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Qualcomm® SA8295P | 3.266 ms | 0 - 45 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 0.917 ms | 0 - 40 MB | NPU |
| Real-ESRGAN-General-x4v3 | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 2.421 ms | 0 - 148 MB | NPU |
License
- The license for the original implementation of Real-ESRGAN-General-x4v3 can be found here.
References
- Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data
- 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.
