Hugging Face's logo

qualcomm
/
EfficientFormer

Image Classification
PyTorch
bu_auto
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
d3ff875
·
verified ·
1 Parent(s): 8c6ca67

v0.60.0

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.60.0 for changelog.

Files changed (2) hide show
  1. README.md +89 -81
  2. release_assets.json +6 -6
README.md CHANGED
@@ -15,7 +15,7 @@ pipeline_tag: image-classification
15
  EfficientFormer is a vision transformer model that can classify images from the Imagenet dataset.
16
 
17
  This is based on the implementation of EfficientFormer found [here](https://github.com/snap-research/EfficientFormer).
18
- 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/v0.59.0/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).
19
 
20
  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.
21
 
@@ -28,111 +28,119 @@ Below are pre-exported model assets ready for deployment.
28
 
29
  | Runtime | Precision | Chipset | SDK Versions | Download |
30
  |---|---|---|---|---|
31
- | ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.59.0/efficientformer-onnx-float.zip)
32
- | ONNX | w8a16 | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.59.0/efficientformer-onnx-w8a16.zip)
33
- | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.59.0/efficientformer-qnn_dlc-float.zip)
34
- | QNN_DLC | w8a16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.59.0/efficientformer-qnn_dlc-w8a16.zip)
35
- | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.59.0/efficientformer-tflite-float.zip)
36
 
37
  For more device-specific assets and performance metrics, visit **[EfficientFormer on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/efficientformer)**.
38
 
39
 
40
  ### Option 2: Export with Custom Configurations
41
 
42
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/efficientformer) Python library to compile and export the model with your own:
43
  - Custom weights (e.g., fine-tuned checkpoints)
44
  - Custom input shapes
45
  - Target device and runtime configurations
46
 
47
  This option is ideal if you need to customize the model beyond the default configuration provided here.
48
 
49
- See our repository for [EfficientFormer on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.59.0/src/qai_hub_models/models/efficientformer) for usage instructions.
50
 
51
  ## Model Details
52
 
53
  **Model Type:** Model_use_case.image_classification
54
 
55
  **Model Stats:**
56
- - Model checkpoint: efficientformer_l1_300d
57
  - Input resolution: 224x224
58
- - Number of parameters: 12.3M
59
  - Model size (float): 46.9 MB
60
  - Model size (w8a16): 12.2 MB
 
61
 
62
  ## Performance Summary
63
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
64
  |---|---|---|---|---|---|---
65
- | EfficientFormer | ONNX | float | Snapdragon® X2 Elite | 0.636 ms | 2 - 2 MB | NPU
66
- | EfficientFormer | ONNX | float | Snapdragon® X Elite | 1.382 ms | 25 - 25 MB | NPU
67
- | EfficientFormer | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.922 ms | 0 - 80 MB | NPU
68
- | EfficientFormer | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 5.091 ms | 1 - 83 MB | NPU
69
- | EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.318 ms | 0 - 137 MB | NPU
70
- | EfficientFormer | ONNX | float | Qualcomm® QCS8450 | 5.091 ms | 1 - 83 MB | NPU
71
- | EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1.765 ms | 1 - 3 MB | NPU
72
- | EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.382 ms | 25 - 25 MB | NPU
73
- | EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 0.695 ms | 0 - 44 MB | NPU
74
- | EfficientFormer | ONNX | float | Snapdragon® 8 Elite Mobile | 0.695 ms | 0 - 44 MB | NPU
75
- | EfficientFormer | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.601 ms | 0 - 44 MB | NPU
76
- | EfficientFormer | ONNX | w8a16 | Snapdragon® X2 Elite | 0.564 ms | 1 - 1 MB | NPU
77
- | EfficientFormer | ONNX | w8a16 | Snapdragon® X Elite | 1.371 ms | 13 - 13 MB | NPU
78
- | EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.896 ms | 0 - 91 MB | NPU
79
- | EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 1.783 ms | 0 - 92 MB | NPU
80
- | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 5.14 ms | 0 - 3 MB | NPU
81
- | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.33 ms | 0 - 16 MB | NPU
82
- | EfficientFormer | ONNX | w8a16 | Qualcomm® QCS8450 | 1.783 ms | 0 - 92 MB | NPU
83
- | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 1.914 ms | 0 - 3 MB | NPU
84
- | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 1.371 ms | 13 - 13 MB | NPU
85
- | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 0.604 ms | 0 - 61 MB | NPU
86
- | EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 0.604 ms | 0 - 61 MB | NPU
87
- | EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.53 ms | 0 - 71 MB | NPU
88
- | EfficientFormer | QNN_DLC | float | Snapdragon® X2 Elite | 0.872 ms | 1 - 1 MB | NPU
89
- | EfficientFormer | QNN_DLC | float | Snapdragon® X Elite | 1.677 ms | 1 - 1 MB | NPU
90
- | EfficientFormer | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.039 ms | 0 - 77 MB | NPU
91
- | EfficientFormer | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 5.579 ms | 0 - 78 MB | NPU
92
- | EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8275 | 4.922 ms | 1 - 39 MB | NPU
93
- | EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.475 ms | 1 - 2 MB | NPU
94
- | EfficientFormer | QNN_DLC | float | Qualcomm® SA8775P | 2.059 ms | 1 - 43 MB | NPU
95
- | EfficientFormer | QNN_DLC | float | Qualcomm® SA8650P | 2.059 ms | 1 - 43 MB | NPU
96
- | EfficientFormer | QNN_DLC | float | Qualcomm® SA8255P | 2.059 ms | 1 - 43 MB | NPU
97
- | EfficientFormer | QNN_DLC | float | Qualcomm® QCS8450 | 5.579 ms | 0 - 78 MB | NPU
98
- | EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 1.9 ms | 3 - 5 MB | NPU
99
- | EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.677 ms | 1 - 1 MB | NPU
100
- | EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 0.786 ms | 0 - 44 MB | NPU
101
- | EfficientFormer | QNN_DLC | float | Qualcomm® SA7255P | 4.922 ms | 1 - 39 MB | NPU
102
- | EfficientFormer | QNN_DLC | float | Qualcomm® SA8295P | 4.008 ms | 0 - 39 MB | NPU
103
- | EfficientFormer | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 0.786 ms | 0 - 44 MB | NPU
104
- | EfficientFormer | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.645 ms | 1 - 45 MB | NPU
105
- | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 0.864 ms | 0 - 0 MB | NPU
106
- | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® X Elite | 1.785 ms | 0 - 0 MB | NPU
107
- | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 1.086 ms | 0 - 80 MB | NPU
108
- | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8275 | 3.226 ms | 0 - 57 MB | NPU
109
- | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.553 ms | 0 - 2 MB | NPU
110
- | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.908 ms | 0 - 59 MB | NPU
111
- | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8650P | 1.908 ms | 0 - 59 MB | NPU
112
- | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8255P | 1.908 ms | 0 - 59 MB | NPU
113
- | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 1.734 ms | 0 - 2 MB | NPU
114
- | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 1.785 ms | 0 - 0 MB | NPU
115
- | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 6.989 ms | 0 - 178 MB | NPU
116
- | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 1.696 ms | 0 - 62 MB | NPU
117
- | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 0.708 ms | 0 - 51 MB | NPU
118
- | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA7255P | 3.226 ms | 0 - 57 MB | NPU
119
- | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 0.708 ms | 0 - 51 MB | NPU
120
- | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.618 ms | 0 - 66 MB | NPU
121
- | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1.696 ms | 0 - 62 MB | NPU
122
- | EfficientFormer | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.037 ms | 0 - 98 MB | NPU
123
- | EfficientFormer | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 5.535 ms | 0 - 96 MB | NPU
124
- | EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ QCS8275 | 4.864 ms | 0 - 52 MB | NPU
125
- | EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.477 ms | 0 - 6 MB | NPU
126
- | EfficientFormer | TFLITE | float | Qualcomm® SA8775P | 2.065 ms | 0 - 55 MB | NPU
127
- | EfficientFormer | TFLITE | float | Qualcomm® SA8650P | 2.065 ms | 0 - 55 MB | NPU
128
- | EfficientFormer | TFLITE | float | Qualcomm® SA8255P | 2.065 ms | 0 - 55 MB | NPU
129
- | EfficientFormer | TFLITE | float | Qualcomm® QCS8450 | 5.535 ms | 0 - 96 MB | NPU
130
- | EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 1.872 ms | 0 - 27 MB | NPU
131
- | EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 0.767 ms | 0 - 58 MB | NPU
132
- | EfficientFormer | TFLITE | float | Qualcomm® SA7255P | 4.864 ms | 0 - 52 MB | NPU
133
- | EfficientFormer | TFLITE | float | Qualcomm® SA8295P | 3.977 ms | 0 - 48 MB | NPU
134
- | EfficientFormer | TFLITE | float | Snapdragon® 8 Elite Mobile | 0.767 ms | 0 - 58 MB | NPU
135
- | EfficientFormer | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.652 ms | 0 - 53 MB | NPU
 
 
 
 
 
 
 
 
136
 
137
  ## License
138
  * The license for the original implementation of EfficientFormer can be found
 
15
  EfficientFormer is a vision transformer model that can classify images from the Imagenet dataset.
16
 
17
  This is based on the implementation of EfficientFormer found [here](https://github.com/snap-research/EfficientFormer).
18
+ 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/v0.60.0/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).
19
 
20
  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.
21
 
 
28
 
29
  | Runtime | Precision | Chipset | SDK Versions | Download |
30
  |---|---|---|---|---|
31
+ | ONNX | float | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.60.0/efficientformer-onnx-float.zip)
32
+ | ONNX | w8a16 | Universal | QAIRT 2.45, ONNX Runtime 1.27.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.60.0/efficientformer-onnx-w8a16.zip)
33
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.60.0/efficientformer-qnn_dlc-float.zip)
34
+ | QNN_DLC | w8a16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.60.0/efficientformer-qnn_dlc-w8a16.zip)
35
+ | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.60.0/efficientformer-tflite-float.zip)
36
 
37
  For more device-specific assets and performance metrics, visit **[EfficientFormer on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/efficientformer)**.
38
 
39
 
40
  ### Option 2: Export with Custom Configurations
41
 
42
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/efficientformer) Python library to compile and export the model with your own:
43
  - Custom weights (e.g., fine-tuned checkpoints)
44
  - Custom input shapes
45
  - Target device and runtime configurations
46
 
47
  This option is ideal if you need to customize the model beyond the default configuration provided here.
48
 
49
+ See our repository for [EfficientFormer on GitHub](https://github.com/qualcomm/ai-hub-models/blob/v0.60.0/src/qai_hub_models/models/efficientformer) for usage instructions.
50
 
51
  ## Model Details
52
 
53
  **Model Type:** Model_use_case.image_classification
54
 
55
  **Model Stats:**
 
56
  - Input resolution: 224x224
57
+ - Model checkpoint: efficientformer_l1_300d
58
  - Model size (float): 46.9 MB
59
  - Model size (w8a16): 12.2 MB
60
+ - Number of parameters: 12.3M
61
 
62
  ## Performance Summary
63
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
64
  |---|---|---|---|---|---|---
65
+ | EfficientFormer | ONNX | float | Snapdragon® X2 Elite | 0.557 ms | 2 - 2 MB | NPU
66
+ | EfficientFormer | ONNX | float | Snapdragon® X Elite | 1.212 ms | 24 - 24 MB | NPU
67
+ | EfficientFormer | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 0.836 ms | 0 - 76 MB | NPU
68
+ | EfficientFormer | ONNX | float | Snapdragon® 8 Gen 1 Mobile | 4.97 ms | 1 - 80 MB | NPU
69
+ | EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ IQ-8275 | 1.603 ms | 1 - 4 MB | NPU
70
+ | EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.15 ms | 0 - 29 MB | NPU
71
+ | EfficientFormer | ONNX | float | Qualcomm® QCS8450 | 4.97 ms | 1 - 80 MB | NPU
72
+ | EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ IQ-9075 | 1.592 ms | 1 - 3 MB | NPU
73
+ | EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.212 ms | 24 - 24 MB | NPU
74
+ | EfficientFormer | ONNX | float | Qualcomm® Dragonwing™ Q-8750 | 0.647 ms | 0 - 41 MB | NPU
75
+ | EfficientFormer | ONNX | float | Snapdragon® 8 Elite Mobile | 0.647 ms | 0 - 41 MB | NPU
76
+ | EfficientFormer | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.575 ms | 0 - 45 MB | NPU
77
+ | EfficientFormer | ONNX | w8a16 | Snapdragon® X2 Elite | 0.516 ms | 1 - 1 MB | NPU
78
+ | EfficientFormer | ONNX | w8a16 | Snapdragon® X Elite | 1.261 ms | 13 - 13 MB | NPU
79
+ | EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.834 ms | 0 - 84 MB | NPU
80
+ | EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Gen 1 Mobile | 1.627 ms | 0 - 86 MB | NPU
81
+ | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS6490 | 3.147 ms | 0 - 3 MB | NPU
82
+ | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 1.236 ms | 0 - 4 MB | NPU
83
+ | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.225 ms | 0 - 107 MB | NPU
84
+ | EfficientFormer | ONNX | w8a16 | Qualcomm® QCS8450 | 1.627 ms | 0 - 86 MB | NPU
85
+ | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 1.384 ms | 0 - 3 MB | NPU
86
+ | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 1.261 ms | 13 - 13 MB | NPU
87
+ | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 5.924 ms | 0 - 67 MB | NPU
88
+ | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 1.361 ms | 0 - 64 MB | NPU
89
+ | EfficientFormer | ONNX | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 0.574 ms | 0 - 58 MB | NPU
90
+ | EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 0.574 ms | 0 - 58 MB | NPU
91
+ | EfficientFormer | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.484 ms | 0 - 65 MB | NPU
92
+ | EfficientFormer | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1.361 ms | 0 - 64 MB | NPU
93
+ | EfficientFormer | QNN_DLC | float | Snapdragon® X2 Elite | 0.928 ms | 1 - 1 MB | NPU
94
+ | EfficientFormer | QNN_DLC | float | Snapdragon® X Elite | 1.68 ms | 1 - 1 MB | NPU
95
+ | EfficientFormer | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 1.032 ms | 0 - 76 MB | NPU
96
+ | EfficientFormer | QNN_DLC | float | Snapdragon® 8 Gen 1 Mobile | 5.59 ms | 0 - 80 MB | NPU
97
+ | EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 1.848 ms | 1 - 4 MB | NPU
98
+ | EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-8275 | 4.869 ms | 1 - 41 MB | NPU
99
+ | EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.485 ms | 1 - 98 MB | NPU
100
+ | EfficientFormer | QNN_DLC | float | Qualcomm® SA8775P | 2.058 ms | 0 - 42 MB | NPU
101
+ | EfficientFormer | QNN_DLC | float | Qualcomm® SA8650P | 2.058 ms | 0 - 42 MB | NPU
102
+ | EfficientFormer | QNN_DLC | float | Qualcomm® SA8255P | 2.058 ms | 0 - 42 MB | NPU
103
+ | EfficientFormer | QNN_DLC | float | Qualcomm® QCS8450 | 5.59 ms | 0 - 80 MB | NPU
104
+ | EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-9075 | 1.903 ms | 1 - 3 MB | NPU
105
+ | EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ IQ-X7181 | 1.68 ms | 1 - 1 MB | NPU
106
+ | EfficientFormer | QNN_DLC | float | Qualcomm® Dragonwing™ Q-8750 | 0.792 ms | 0 - 40 MB | NPU
107
+ | EfficientFormer | QNN_DLC | float | Qualcomm® SA7255P | 4.869 ms | 1 - 41 MB | NPU
108
+ | EfficientFormer | QNN_DLC | float | Qualcomm® SA8295P | 3.983 ms | 0 - 39 MB | NPU
109
+ | EfficientFormer | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 0.792 ms | 0 - 40 MB | NPU
110
+ | EfficientFormer | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.654 ms | 1 - 43 MB | NPU
111
+ | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 0.803 ms | 0 - 0 MB | NPU
112
+ | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® X Elite | 1.686 ms | 0 - 0 MB | NPU
113
+ | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 0.99 ms | 0 - 70 MB | NPU
114
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 1.441 ms | 0 - 3 MB | NPU
115
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-8275 | 3.082 ms | 0 - 55 MB | NPU
116
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.471 ms | 0 - 2 MB | NPU
117
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8775P | 1.774 ms | 0 - 58 MB | NPU
118
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8650P | 1.774 ms | 0 - 58 MB | NPU
119
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA8255P | 1.774 ms | 0 - 58 MB | NPU
120
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-9075 | 1.717 ms | 0 - 2 MB | NPU
121
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ IQ-X7181 | 1.686 ms | 0 - 0 MB | NPU
122
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-6690 | 6.866 ms | 0 - 171 MB | NPU
123
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-7790 | 1.633 ms | 0 - 59 MB | NPU
124
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® Dragonwing™ Q-8750 | 0.669 ms | 0 - 50 MB | NPU
125
+ | EfficientFormer | QNN_DLC | w8a16 | Qualcomm® SA7255P | 3.082 ms | 0 - 55 MB | NPU
126
+ | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 0.669 ms | 0 - 50 MB | NPU
127
+ | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 0.566 ms | 0 - 56 MB | NPU
128
+ | EfficientFormer | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 1.633 ms | 0 - 59 MB | NPU
129
+ | EfficientFormer | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 1.04 ms | 0 - 96 MB | NPU
130
+ | EfficientFormer | TFLITE | float | Snapdragon® 8 Gen 1 Mobile | 5.55 ms | 0 - 97 MB | NPU
131
+ | EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 1.832 ms | 0 - 28 MB | NPU
132
+ | EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ IQ-8275 | 4.856 ms | 0 - 52 MB | NPU
133
+ | EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ QCS8550 (Proxy) | 1.474 ms | 0 - 3 MB | NPU
134
+ | EfficientFormer | TFLITE | float | Qualcomm® SA8775P | 2.055 ms | 0 - 55 MB | NPU
135
+ | EfficientFormer | TFLITE | float | Qualcomm® SA8650P | 2.055 ms | 0 - 55 MB | NPU
136
+ | EfficientFormer | TFLITE | float | Qualcomm® SA8255P | 2.055 ms | 0 - 55 MB | NPU
137
+ | EfficientFormer | TFLITE | float | Qualcomm® QCS8450 | 5.55 ms | 0 - 97 MB | NPU
138
+ | EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ IQ-9075 | 1.904 ms | 0 - 27 MB | NPU
139
+ | EfficientFormer | TFLITE | float | Qualcomm® Dragonwing™ Q-8750 | 0.762 ms | 0 - 49 MB | NPU
140
+ | EfficientFormer | TFLITE | float | Qualcomm® SA7255P | 4.856 ms | 0 - 52 MB | NPU
141
+ | EfficientFormer | TFLITE | float | Qualcomm® SA8295P | 3.969 ms | 0 - 47 MB | NPU
142
+ | EfficientFormer | TFLITE | float | Snapdragon® 8 Elite Mobile | 0.762 ms | 0 - 49 MB | NPU
143
+ | EfficientFormer | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 0.653 ms | 0 - 53 MB | NPU
144
 
145
  ## License
146
  * The license for the original implementation of EfficientFormer can be found
release_assets.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "version": "0.59.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
@@ -8,19 +8,19 @@
8
  "qairt": "2.45.0.260326154327",
9
  "onnx_runtime": "1.27.1"
10
  },
11
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.59.0/efficientformer-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
  "qairt": "2.45.0.260326154327"
16
  },
17
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.59.0/efficientformer-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
  "qairt": "2.45.0.260326154327"
22
  },
23
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.59.0/efficientformer-tflite-float.zip"
24
  }
25
  }
26
  },
@@ -31,13 +31,13 @@
31
  "qairt": "2.45.0.260326154327",
32
  "onnx_runtime": "1.27.1"
33
  },
34
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.59.0/efficientformer-onnx-w8a16.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.45.0.260326154327"
39
  },
40
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.59.0/efficientformer-qnn_dlc-w8a16.zip"
41
  }
42
  }
43
  }
 
1
  {
2
+ "version": "0.60.0",
3
  "precisions": {
4
  "float": {
5
  "universal_assets": {
 
8
  "qairt": "2.45.0.260326154327",
9
  "onnx_runtime": "1.27.1"
10
  },
11
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.60.0/efficientformer-onnx-float.zip"
12
  },
13
  "qnn_dlc": {
14
  "tool_versions": {
15
  "qairt": "2.45.0.260326154327"
16
  },
17
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.60.0/efficientformer-qnn_dlc-float.zip"
18
  },
19
  "tflite": {
20
  "tool_versions": {
21
  "qairt": "2.45.0.260326154327"
22
  },
23
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.60.0/efficientformer-tflite-float.zip"
24
  }
25
  }
26
  },
 
31
  "qairt": "2.45.0.260326154327",
32
  "onnx_runtime": "1.27.1"
33
  },
34
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.60.0/efficientformer-onnx-w8a16.zip"
35
  },
36
  "qnn_dlc": {
37
  "tool_versions": {
38
  "qairt": "2.45.0.260326154327"
39
  },
40
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/efficientformer/releases/v0.60.0/efficientformer-qnn_dlc-w8a16.zip"
41
  }
42
  }
43
  }