Hugging Face's logo

qualcomm
/
EfficientFormer

Image Classification
PyTorch
bu_auto
android
Model card Files
xet
 Community
EfficientFormer
File size: 11,123 Bytes
f482057
 
 
 
77634dd
f482057
 
 
 
 
 
 
1c564d0
f482057
 
 
1c564d0
e984d92
1c564d0
 
 
 
 
 
 
 
 
 
 
 
e984d92
 
 
 
 
1c564d0
 
 
 
 
 
e984d92
1c564d0
 
 
 
 
 
e984d92
1c564d0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e984d92
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
f482057
 
 
 
 
 
 
 
 
 
 
 
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
---
library_name: pytorch
license: other
tags:
- bu_auto
- android
pipeline_tag: image-classification

---

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

# EfficientFormer: Optimized for Qualcomm Devices

EfficientFormer is a vision transformer model that can classify images from the Imagenet dataset.

This is based on the implementation of EfficientFormer found [here](https://github.com/snap-research/EfficientFormer).
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/efficientformer) 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/efficientformer/releases/v0.58.0/efficientformer-onnx-float.zip)
| ONNX | w8a16 | Universal | QAIRT 2.45, ONNX Runtime 1.25.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.58.0/efficientformer-onnx-w8a16.zip)
| QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.58.0/efficientformer-qnn_dlc-float.zip)
| QNN_DLC | w8a16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.58.0/efficientformer-qnn_dlc-w8a16.zip)
| TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.58.0/efficientformer-tflite-float.zip)

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


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

## Model Details

**Model Type:** Model_use_case.image_classification

**Model Stats:**
- Model checkpoint: efficientformer_l1_300d
- Input resolution: 224x224
- Number of parameters: 12.3M
- Model size (float): 46.9 MB
- Model size (w8a16): 12.2 MB

## Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
|---|---|---|---|---|---|---
| EfficientFormer | ONNX | float | Snapdragon® X2 Elite | 0.635 ms | 2 - 2 MB | NPU
| EfficientFormer | ONNX | float | Snapdragon® X Elite | 1.376 ms | 24 - 24 MB | NPU
| EfficientFormer | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.915 ms | 0 - 79 MB | NPU
| EfficientFormer | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 5.095 ms | 1 - 83 MB | NPU
| EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.324 ms | 0 - 39 MB | NPU
| EfficientFormer | ONNX | float | Qualcomm® QCS8450 | 5.095 ms | 1 - 83 MB | NPU
| EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1.759 ms | 1 - 3 MB | NPU
| EfficientFormer | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.612 ms | 0 - 45 MB | NPU
| EfficientFormer | ONNX | float | Snapdragon® 8 Elite Mobile | 0.69 ms | 0 - 44 MB | NPU
| EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 0.69 ms | 0 - 44 MB | NPU
| EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.376 ms | 24 - 24 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® X2 Elite | 0.567 ms | 1 - 1 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® X Elite | 1.378 ms | 12 - 12 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.898 ms | 0 - 95 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 2.216 ms | 0 - 87 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 5.752 ms | 1 - 3 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.319 ms | 0 - 3 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® QCS8450 | 2.216 ms | 0 - 87 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 1.573 ms | 0 - 3 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1.469 ms | 0 - 70 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.532 ms | 0 - 72 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 6.234 ms | 0 - 187 MB | NPU
| EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 0.609 ms | 0 - 61 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 1.469 ms | 0 - 70 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 0.609 ms | 0 - 61 MB | NPU
| EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 1.378 ms | 12 - 12 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® X2 Elite | 0.92 ms | 1 - 1 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® X Elite | 1.675 ms | 1 - 1 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.036 ms | 0 - 75 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 5.577 ms | 0 - 77 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® QCS8275 | 4.896 ms | 0 - 40 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.479 ms | 1 - 2 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® SA8775P | 2.068 ms | 1 - 43 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® SA8650P | 2.068 ms | 1 - 43 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® SA8255P | 2.068 ms | 1 - 43 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® QCS8450 | 5.577 ms | 0 - 77 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 2.108 ms | 1 - 3 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.65 ms | 1 - 45 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® SA7255P | 4.896 ms | 0 - 40 MB | NPU
| EfficientFormer | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 0.769 ms | 0 - 44 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® SA8295P | 3.989 ms | 1 - 39 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 0.769 ms | 0 - 44 MB | NPU
| EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.675 ms | 1 - 1 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 0.885 ms | 0 - 0 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® X Elite | 1.781 ms | 0 - 0 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 1.075 ms | 0 - 80 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® QCS8275 | 3.241 ms | 0 - 58 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.563 ms | 0 - 2 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.892 ms | 0 - 59 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8650P | 1.892 ms | 0 - 59 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8255P | 1.892 ms | 0 - 59 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 1.735 ms | 2 - 4 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1.692 ms | 0 - 62 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.615 ms | 0 - 65 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 6.987 ms | 0 - 178 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA7255P | 3.241 ms | 0 - 58 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 0.705 ms | 0 - 52 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 1.692 ms | 0 - 62 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 0.705 ms | 0 - 52 MB | NPU
| EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 1.781 ms | 0 - 0 MB | NPU
| EfficientFormer | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.031 ms | 0 - 88 MB | NPU
| EfficientFormer | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 5.541 ms | 0 - 96 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® QCS8275 | 4.852 ms | 0 - 52 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.469 ms | 0 - 29 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® SA8775P | 2.066 ms | 0 - 53 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® SA8650P | 2.066 ms | 0 - 53 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® SA8255P | 2.066 ms | 0 - 53 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® QCS8450 | 5.541 ms | 0 - 96 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 1.886 ms | 0 - 27 MB | NPU
| EfficientFormer | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.655 ms | 0 - 53 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® SA7255P | 4.852 ms | 0 - 52 MB | NPU
| EfficientFormer | TFLITE | float | Snapdragon® 8 Elite Mobile | 0.779 ms | 0 - 56 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® SA8295P | 4.018 ms | 0 - 47 MB | NPU
| EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 0.779 ms | 0 - 56 MB | NPU

## License
* The license for the original implementation of EfficientFormer can be found
  [here](https://github.com/snap-research/EfficientFormer?tab=License-1-ov-file#readme).

## References
* [Rethinking Vision Transformers for MobileNet Size and Speed](https://arxiv.org/abs/2212.08059)
* [Source Model Implementation](https://github.com/snap-research/EfficientFormer)

## 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).