Hugging Face's logo

qualcomm
/
QuickSRNetLarge

Image-to-Image
PyTorch
android
Model card Files
xet
 Community
QuickSRNetLarge
File size: 13,344 Bytes
fe32929
 
 
 
 
3b79ef9
fe32929
 
 
9f647b5
fe32929
cdd46c4
950b6b2
fe32929
 
cdd46c4
6a8b0a4
cdd46c4
 
 
 
 
 
 
 
 
 
 
 
6a8b0a4
 
 
 
 
 
cdd46c4
 
 
 
 
 
6a8b0a4
cdd46c4
 
 
 
 
 
6a8b0a4
cdd46c4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
6a8b0a4
 
 
 
 
 
 
 
f8dfd30
6a8b0a4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f8dfd30
6a8b0a4
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f8dfd30
6a8b0a4
 
 
 
 
 
f8dfd30
 
6a8b0a4
 
 
 
 
 
 
 
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
168
169
170
171
172
173
---
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/v0.58.0/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/v0.58.0/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/v0.58.0/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/v0.58.0/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/v0.58.0/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/v0.58.0/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/v0.58.0/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/v0.58.0/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/v0.58.0/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 | 1.203 ms | 8 - 8 MB | NPU
| QuickSRNetLarge | ONNX | float | Snapdragon® X Elite | 1.996 ms | 8 - 8 MB | NPU
| QuickSRNetLarge | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 1.436 ms | 0 - 38 MB | NPU
| QuickSRNetLarge | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 3.451 ms | 0 - 45 MB | NPU
| QuickSRNetLarge | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.907 ms | 0 - 2 MB | NPU
| QuickSRNetLarge | ONNX | float | Qualcomm® QCS8450 | 3.451 ms | 0 - 45 MB | NPU
| QuickSRNetLarge | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 3.326 ms | 8 - 11 MB | NPU
| QuickSRNetLarge | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.94 ms | 0 - 23 MB | NPU
| QuickSRNetLarge | ONNX | float | Snapdragon® 8 Elite Mobile | 1.2 ms | 0 - 20 MB | NPU
| QuickSRNetLarge | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 1.2 ms | 0 - 20 MB | NPU
| QuickSRNetLarge | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.996 ms | 8 - 8 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® X2 Elite | 0.353 ms | 3 - 3 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® X Elite | 0.785 ms | 3 - 3 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.523 ms | 0 - 37 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 1.071 ms | 0 - 41 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 3.141 ms | 0 - 3 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.777 ms | 0 - 2 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® QCS8450 | 1.071 ms | 0 - 41 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 0.384 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 1.106 ms | 0 - 25 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.329 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.909 ms | 0 - 3 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 9.034 ms | 2 - 140 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 1.106 ms | 0 - 25 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.384 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 0.785 ms | 3 - 3 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® X2 Elite | 1.245 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® X Elite | 1.997 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.286 ms | 0 - 37 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 3.315 ms | 0 - 43 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® QCS8275 | 11.857 ms | 0 - 22 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.808 ms | 0 - 8 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® SA8775P | 3.405 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® SA8650P | 3.405 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® SA8255P | 3.405 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® QCS8450 | 3.315 ms | 0 - 43 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 3.24 ms | 2 - 7 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.869 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® SA7255P | 11.857 ms | 0 - 22 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 1.032 ms | 0 - 23 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® SA8295P | 3.944 ms | 0 - 20 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 1.032 ms | 0 - 23 MB | NPU
| QuickSRNetLarge | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.997 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.397 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.769 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.439 ms | 0 - 33 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.816 ms | 0 - 39 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 3.086 ms | 0 - 2 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® QCS8275 | 2.091 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.647 ms | 0 - 1 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.86 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® SA8650P | 0.86 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® SA8255P | 0.86 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 0.816 ms | 0 - 39 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 0.373 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.905 ms | 0 - 22 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.273 ms | 0 - 25 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.753 ms | 2 - 4 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 8.33 ms | 0 - 136 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® SA7255P | 2.091 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® SA8295P | 1.364 ms | 0 - 21 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 0.905 ms | 0 - 22 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.373 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 0.769 ms | 0 - 0 MB | NPU
| QuickSRNetLarge | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.449 ms | 0 - 39 MB | NPU
| QuickSRNetLarge | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 3.557 ms | 0 - 44 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® QCS8275 | 12.202 ms | 3 - 27 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 2.138 ms | 0 - 2 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® SA8775P | 3.8 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® SA8650P | 3.8 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® SA8255P | 3.8 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® QCS8450 | 3.557 ms | 0 - 44 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 3.62 ms | 3 - 9 MB | NPU
| QuickSRNetLarge | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.879 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® SA7255P | 12.202 ms | 3 - 27 MB | NPU
| QuickSRNetLarge | TFLITE | float | Snapdragon® 8 Elite Mobile | 1.078 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® SA8295P | 4.212 ms | 0 - 22 MB | NPU
| QuickSRNetLarge | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 1.078 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.49 ms | 0 - 34 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.91 ms | 0 - 40 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 2.946 ms | 0 - 3 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® QCS8275 | 2.179 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.785 ms | 0 - 1 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® SA8775P | 1.004 ms | 0 - 27 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® SA8650P | 1.004 ms | 0 - 27 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® SA8255P | 1.004 ms | 0 - 27 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® QCS8450 | 0.91 ms | 0 - 40 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 0.386 ms | 0 - 29 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.973 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.3 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.919 ms | 0 - 3 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 8.548 ms | 0 - 136 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® SA7255P | 2.179 ms | 0 - 26 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® SA8295P | 1.469 ms | 0 - 23 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 0.973 ms | 0 - 24 MB | NPU
| QuickSRNetLarge | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 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).