Hugging Face's logo

qualcomm
/
QuickSRNetSmall

Image-to-Image
PyTorch
android
Model card Files
xet
 Community
QuickSRNetSmall
File size: 13,351 Bytes
79f470b
 
 
 
 
f8da907
79f470b
 
 
a71ce34
79f470b
c95eb37
e815d88
79f470b
 
c95eb37
c3dc8f5
c95eb37
 
 
 
 
 
 
 
 
 
 
 
c3dc8f5
 
 
 
 
 
c95eb37
 
 
 
 
 
c3dc8f5
c95eb37
 
 
 
 
 
c3dc8f5
c95eb37
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c3dc8f5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
8e8123f
79f470b
3bcc01a
9c155b2
8e8123f
79f470b
 
 
 
 
83eb19d
79f470b
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/quicksrnetsmall/web-assets/model_demo.png)

# QuickSRNetSmall: Optimized for Qualcomm Devices

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

This is based on the implementation of QuickSRNetSmall 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/quicksrnetsmall) 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/quicksrnetsmall/releases/v0.58.0/quicksrnetsmall-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/quicksrnetsmall/releases/v0.58.0/quicksrnetsmall-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/quicksrnetsmall/releases/v0.58.0/quicksrnetsmall-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/quicksrnetsmall/releases/v0.58.0/quicksrnetsmall-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/quicksrnetsmall/releases/v0.58.0/quicksrnetsmall-tflite-float.zip)
| TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/quicksrnetsmall/releases/v0.58.0/quicksrnetsmall-tflite-w8a8.zip)

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


### 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/quicksrnetsmall) 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 [QuickSRNetSmall on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.58.0/src/qai_hub_models/models/quicksrnetsmall) for usage instructions.

## Model Details

**Model Type:** Model_use_case.super_resolution

**Model Stats:**
- Model checkpoint: quicksrnet_small_3x_checkpoint
- Input resolution: 128x128
- Number of parameters: 33.3K
- Model size (float): 133 KB
- Model size (w8a8): 41.7 KB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| QuickSRNetSmall | ONNX | float | Snapdragon® X2 Elite | 0.408 ms | 8 - 8 MB | NPU
| QuickSRNetSmall | ONNX | float | Snapdragon® X Elite | 0.849 ms | 8 - 8 MB | NPU
| QuickSRNetSmall | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.544 ms | 0 - 28 MB | NPU
| QuickSRNetSmall | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 1.201 ms | 0 - 34 MB | NPU
| QuickSRNetSmall | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.838 ms | 0 - 7 MB | NPU
| QuickSRNetSmall | ONNX | float | Qualcomm® QCS8450 | 1.201 ms | 0 - 34 MB | NPU
| QuickSRNetSmall | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1.107 ms | 8 - 11 MB | NPU
| QuickSRNetSmall | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.398 ms | 0 - 19 MB | NPU
| QuickSRNetSmall | ONNX | float | Snapdragon® 8 Elite Mobile | 0.434 ms | 0 - 17 MB | NPU
| QuickSRNetSmall | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 0.434 ms | 0 - 17 MB | NPU
| QuickSRNetSmall | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 0.849 ms | 8 - 8 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Snapdragon® X2 Elite | 0.19 ms | 3 - 3 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Snapdragon® X Elite | 0.493 ms | 3 - 3 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.332 ms | 0 - 28 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.572 ms | 0 - 29 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 2.113 ms | 0 - 3 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.496 ms | 0 - 4 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Qualcomm® QCS8450 | 0.572 ms | 0 - 29 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 0.236 ms | 0 - 19 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.51 ms | 0 - 18 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.211 ms | 0 - 20 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.665 ms | 0 - 3 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 1.795 ms | 0 - 132 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 0.51 ms | 0 - 18 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.236 ms | 0 - 19 MB | NPU
| QuickSRNetSmall | ONNX | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 0.493 ms | 3 - 3 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Snapdragon® X2 Elite | 0.461 ms | 0 - 0 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Snapdragon® X Elite | 0.832 ms | 0 - 0 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 0.444 ms | 0 - 28 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 1.044 ms | 0 - 34 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Qualcomm® QCS8275 | 1.924 ms | 0 - 20 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.736 ms | 0 - 1 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Qualcomm® SA8775P | 1.062 ms | 0 - 22 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Qualcomm® SA8650P | 1.062 ms | 0 - 22 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Qualcomm® SA8255P | 1.062 ms | 0 - 22 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Qualcomm® QCS8450 | 1.044 ms | 0 - 34 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 0.956 ms | 2 - 7 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.328 ms | 0 - 22 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Qualcomm® SA7255P | 1.924 ms | 0 - 20 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 0.349 ms | 0 - 20 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Qualcomm® SA8295P | 1.4 ms | 0 - 17 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 0.349 ms | 0 - 20 MB | NPU
| QuickSRNetSmall | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 0.832 ms | 0 - 0 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 0.239 ms | 0 - 0 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Snapdragon® X Elite | 0.46 ms | 0 - 0 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.222 ms | 0 - 28 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.425 ms | 0 - 32 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 1.3 ms | 0 - 2 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® QCS8275 | 0.837 ms | 0 - 19 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.349 ms | 0 - 2 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® SA8775P | 0.537 ms | 0 - 21 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® SA8650P | 0.537 ms | 0 - 21 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® SA8255P | 0.537 ms | 0 - 21 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® QCS8450 | 0.425 ms | 0 - 32 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 0.208 ms | 0 - 20 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.362 ms | 0 - 18 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.163 ms | 0 - 21 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.474 ms | 2 - 4 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 1.261 ms | 0 - 16 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® SA7255P | 0.837 ms | 0 - 19 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® SA8295P | 0.711 ms | 0 - 17 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 0.362 ms | 0 - 18 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.208 ms | 0 - 20 MB | NPU
| QuickSRNetSmall | QNN_DLC | w8a8 | Qualcomm® Dragonwing™ IQ-X7181 | 0.46 ms | 0 - 0 MB | NPU
| QuickSRNetSmall | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 0.611 ms | 0 - 28 MB | NPU
| QuickSRNetSmall | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 1.316 ms | 0 - 29 MB | NPU
| QuickSRNetSmall | TFLITE | float | Qualcomm® QCS8275 | 2.376 ms | 0 - 21 MB | NPU
| QuickSRNetSmall | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.004 ms | 0 - 2 MB | NPU
| QuickSRNetSmall | TFLITE | float | Qualcomm® SA8775P | 1.412 ms | 0 - 22 MB | NPU
| QuickSRNetSmall | TFLITE | float | Qualcomm® SA8650P | 1.412 ms | 0 - 22 MB | NPU
| QuickSRNetSmall | TFLITE | float | Qualcomm® SA8255P | 1.412 ms | 0 - 22 MB | NPU
| QuickSRNetSmall | TFLITE | float | Qualcomm® QCS8450 | 1.316 ms | 0 - 29 MB | NPU
| QuickSRNetSmall | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 1.272 ms | 3 - 8 MB | NPU
| QuickSRNetSmall | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.395 ms | 0 - 21 MB | NPU
| QuickSRNetSmall | TFLITE | float | Qualcomm® SA7255P | 2.376 ms | 0 - 21 MB | NPU
| QuickSRNetSmall | TFLITE | float | Snapdragon® 8 Elite Mobile | 0.445 ms | 0 - 20 MB | NPU
| QuickSRNetSmall | TFLITE | float | Qualcomm® SA8295P | 1.655 ms | 0 - 17 MB | NPU
| QuickSRNetSmall | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 0.445 ms | 0 - 20 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 0.275 ms | 0 - 28 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Snapdragon® 8 Gen 1 Mobile | 0.468 ms | 0 - 33 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS6490 | 1.19 ms | 0 - 2 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® QCS8275 | 0.909 ms | 0 - 19 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 0.412 ms | 0 - 4 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® SA8775P | 0.611 ms | 0 - 21 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® SA8650P | 0.611 ms | 0 - 21 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® SA8255P | 0.611 ms | 0 - 21 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® QCS8450 | 0.468 ms | 0 - 33 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 0.216 ms | 0 - 24 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 0.436 ms | 0 - 18 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 0.179 ms | 0 - 21 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® Dragonwing™ IQ-9075 | 0.543 ms | 0 - 3 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-6690 | 1.504 ms | 0 - 17 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® SA7255P | 0.909 ms | 0 - 19 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® SA8295P | 0.802 ms | 0 - 17 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-7790 | 0.436 ms | 0 - 18 MB | NPU
| QuickSRNetSmall | TFLITE | w8a8 | Qualcomm® Dragonwing™ Q-8750 | 0.216 ms | 0 - 24 MB | NPU

## License
* The license for the original implementation of QuickSRNetSmall 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).