Hugging Face's logo

qualcomm
/
QuickSRNetLarge

Image-to-Image
PyTorch
android
Model card Files
xet
 Community
QuickSRNetLarge
File size: 12,434 Bytes
fe32929
 
 
 
 
3b79ef9
fe32929
 
 
9f647b5
fe32929
cdd46c4
950b6b2
fe32929
 
cdd46c4
2f228f6
cdd46c4
 
 
 
 
 
 
 
 
 
 
 
f8dfd30
 
 
 
 
 
cdd46c4
 
 
 
 
 
2f228f6
cdd46c4
 
 
 
 
 
2f228f6
cdd46c4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f8dfd30
 
 
 
 
 
 
 
fb66a2d
f8dfd30
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8a85c0d
fe32929
7a72693
523b94a
8a85c0d
fe32929
 
 
 
 
af5342c
fe32929
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
---
library_name: pytorch
license: other
tags:
- android
pipeline_tag: image-to-image

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/quicksrnetlarge/web-assets/model_demo.png)

# QuickSRNetLarge: Optimized for Qualcomm Devices

QuickSRNet Large is designed for upscaling images on mobile platforms to sharpen in real-time.

This is based on the implementation of QuickSRNetLarge found [here](https://github.com/quic/aimet-model-zoo/tree/develop/aimet_zoo_torch/quicksrnet).
This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/quicksrnetlarge) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary).

Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) 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](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/quicksrnetlarge/releases/v0.56.0/quicksrnetlarge-onnx-float.zip)
| ONNX | w8a8 | Universal | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/quicksrnetlarge/releases/v0.56.0/quicksrnetlarge-onnx-w8a8.zip)
| QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/quicksrnetlarge/releases/v0.56.0/quicksrnetlarge-qnn_dlc-float.zip)
| QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/quicksrnetlarge/releases/v0.56.0/quicksrnetlarge-qnn_dlc-w8a8.zip)
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/quicksrnetlarge/releases/v0.56.0/quicksrnetlarge-tflite-float.zip)
| TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/quicksrnetlarge/releases/v0.56.0/quicksrnetlarge-tflite-w8a8.zip)

For more device-specific assets and performance metrics, visit **[QuickSRNetLarge on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/quicksrnetlarge)**.


### Option 2: Export with Custom Configurations

Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/quicksrnetlarge) 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 [QuickSRNetLarge on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/quicksrnetlarge) for usage instructions.

## Model Details

**Model Type:** Model_use_case.super_resolution

**Model Stats:**
- Model checkpoint: quicksrnet_large_3x_checkpoint
- Input resolution: 128x128
- Number of parameters: 436K
- Model size (float): 1.67 MB
- Model size (w8a8): 462 KB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| QuickSRNetLarge | ONNX | float | Snapdragon® X2 Elite | 1.197 ms | 213 - 213 MB | NPU
| QuickSRNetLarge | ONNX | float | Snapdragon® X Elite | 1.981 ms | 150 - 150 MB | NPU
| QuickSRNetLarge | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.433 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 3.46 ms | 0 - 36 MB | NPU
| QuickSRNetLarge | ONNX | float | Qualcomm® QCS8550 (Proxy) | 1.904 ms | 0 - 5 MB | NPU
| QuickSRNetLarge | ONNX | float | Qualcomm® QCS8450 | 3.46 ms | 0 - 36 MB | NPU
| QuickSRNetLarge | ONNX | float | Snapdragon® 8 Elite Mobile | 1.2 ms | 0 - 20 MB | NPU
| QuickSRNetLarge | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.916 ms | 0 - 17 MB | NPU
| QuickSRNetLarge | ONNX | float | Qualcomm® QCS9075 | 3.523 ms | 7 - 53 MB | NPU
| QuickSRNetLarge | ONNX | float | Qualcomm® QCS8750 | 1.2 ms | 0 - 20 MB | NPU
| QuickSRNetLarge | ONNX | float | Qualcomm® QCS7181 | 1.981 ms | 150 - 150 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® X2 Elite | 0.354 ms | 213 - 213 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® X Elite | 0.779 ms | 181 - 181 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.514 ms | 0 - 29 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.068 ms | 2 - 35 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® QCS6490 | 3.063 ms | 1 - 47 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.78 ms | 0 - 6 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® QCS8450 | 1.068 ms | 2 - 35 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.316 ms | 0 - 21 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.107 ms | 0 - 21 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® QCM6690 | 9.06 ms | 1 - 140 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® QCS9075 | 0.991 ms | 0 - 46 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 0.383 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® QCS7790 | 1.107 ms | 0 - 21 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® QCS8750 | 0.383 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® QCS7181 | 0.779 ms | 181 - 181 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® X2 Elite | 1.245 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® X Elite | 1.971 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.29 ms | 0 - 37 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 3.265 ms | 0 - 43 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® QCS8275 | 11.87 ms | 0 - 23 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 1.805 ms | 0 - 4 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® QCS8450 | 3.265 ms | 0 - 43 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 1.041 ms | 0 - 23 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® SA8295P | 3.915 ms | 0 - 20 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.837 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® SA7255P | 11.87 ms | 0 - 23 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® QCS9075 | 3.363 ms | 0 - 5 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® QCS8750 | 1.041 ms | 0 - 23 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® QCS7181 | 1.971 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.423 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.777 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.442 ms | 0 - 34 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.823 ms | 0 - 35 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 2.736 ms | 0 - 2 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® QCS8275 | 2.121 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.649 ms | 0 - 30 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 0.823 ms | 0 - 35 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.267 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.912 ms | 0 - 22 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 8.322 ms | 0 - 136 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 0.851 ms | 2 - 4 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® SA7255P | 2.121 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 0.374 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® SA8295P | 1.374 ms | 0 - 21 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® QCS7790 | 0.912 ms | 0 - 22 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® QCS8750 | 0.374 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® QCS7181 | 0.777 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.448 ms | 0 - 39 MB | NPU
| QuickSRNetLarge | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 3.499 ms | 0 - 44 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® QCS8275 | 12.183 ms | 3 - 28 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 2.138 ms | 0 - 2 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® SA8775P | 27.916 ms | 3 - 20 MB | GPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® SA8650P | 27.916 ms | 3 - 20 MB | GPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® SA8255P | 27.916 ms | 3 - 20 MB | GPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® QCS8450 | 3.499 ms | 0 - 44 MB | NPU
| QuickSRNetLarge | TFLITE | float | Snapdragon® 8 Elite Mobile | 1.081 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® SA8295P | 4.182 ms | 0 - 22 MB | NPU
| QuickSRNetLarge | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.866 ms | 0 - 28 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® SA7255P | 12.183 ms | 3 - 28 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® QCS9075 | 3.704 ms | 3 - 9 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® QCS8750 | 1.081 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.504 ms | 0 - 34 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.91 ms | 0 - 40 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® QCS6490 | 2.951 ms | 0 - 3 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® QCS8275 | 2.188 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 0.784 ms | 0 - 1 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® SA8775P | 28.64 ms | 4 - 26 MB | GPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® SA8650P | 28.64 ms | 4 - 26 MB | GPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® SA8255P | 28.64 ms | 4 - 26 MB | GPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® QCS8450 | 0.91 ms | 0 - 40 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.287 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.969 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® QCM6690 | 8.555 ms | 0 - 137 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® QCS9075 | 0.978 ms | 0 - 3 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® SA7255P | 2.188 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 0.386 ms | 0 - 29 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® SA8295P | 1.47 ms | 0 - 23 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® QCS7790 | 0.969 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® QCS8750 | 0.386 ms | 0 - 29 MB | NPU

## License
* The license for the original implementation of QuickSRNetLarge can be found
  [here](https://github.com/quic/aimet-model-zoo/blob/develop/LICENSE.md).

## References
* [QuickSRNet: Plain Single-Image Super-Resolution Architecture for Faster Inference on Mobile Platforms](https://arxiv.org/abs/2303.04336)
* [Source Model Implementation](https://github.com/quic/aimet-model-zoo/tree/develop/aimet_zoo_torch/quicksrnet)

## Community
* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
* For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).