Hugging Face's logo

qualcomm
/
Segformer-Base

Image Segmentation
PyTorch
real_time
android
Model card Files
xet
 Community
qaihm-bot commited on
Commit
36811d3
·
verified ·
1 Parent(s): 8edf9fd

v0.49.1

Browse files

See https://github.com/qualcomm/ai-hub-models/releases/v0.49.1 for changelog.

Files changed (1) hide show
  1. README.md +73 -73
README.md CHANGED
@@ -15,7 +15,7 @@ pipeline_tag: image-segmentation
15
  Segformer Base is a machine learning model that predicts masks and classes of objects in an image.
16
 
17
  This is based on the implementation of Segformer-Base found [here](https://github.com/huggingface/transformers/tree/main/src/transformers/models/segformer).
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/main/qai_hub_models/models/segformer_base) 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
 
@@ -28,28 +28,28 @@ Below are pre-exported model assets ready for deployment.
28
 
29
  | Runtime | Precision | Chipset | SDK Versions | Download |
30
  |---|---|---|---|---|
31
- | ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.48.0/segformer_base-onnx-float.zip)
32
- | ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.48.0/segformer_base-onnx-w8a16.zip)
33
- | ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.48.0/segformer_base-onnx-w8a8.zip)
34
- | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.48.0/segformer_base-qnn_dlc-float.zip)
35
- | QNN_DLC | w8a16 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.48.0/segformer_base-qnn_dlc-w8a16.zip)
36
- | QNN_DLC | w8a8 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.48.0/segformer_base-qnn_dlc-w8a8.zip)
37
- | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.48.0/segformer_base-tflite-float.zip)
38
- | TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.48.0/segformer_base-tflite-w8a8.zip)
39
 
40
  For more device-specific assets and performance metrics, visit **[Segformer-Base on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/segformer_base)**.
41
 
42
 
43
  ### Option 2: Export with Custom Configurations
44
 
45
- Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/segformer_base) Python library to compile and export the model with your own:
46
  - Custom weights (e.g., fine-tuned checkpoints)
47
  - Custom input shapes
48
  - Target device and runtime configurations
49
 
50
  This option is ideal if you need to customize the model beyond the default configuration provided here.
51
 
52
- See our repository for [Segformer-Base on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/qai_hub_models/models/segformer_base) for usage instructions.
53
 
54
  ## Model Details
55
 
@@ -67,68 +67,68 @@ See our repository for [Segformer-Base on GitHub](https://github.com/qualcomm/ai
67
  ## Performance Summary
68
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
69
  |---|---|---|---|---|---|---
70
- | Segformer-Base | ONNX | float | Snapdragon® X2 Elite | 72.601 ms | 33 - 33 MB | NPU
71
- | Segformer-Base | ONNX | float | Snapdragon® X Elite | 112.285 ms | 33 - 33 MB | NPU
72
- | Segformer-Base | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 82.369 ms | 26 - 258 MB | NPU
73
- | Segformer-Base | ONNX | float | Qualcomm® QCS8550 (Proxy) | 108.088 ms | 19 - 29 MB | NPU
74
- | Segformer-Base | ONNX | float | Qualcomm® QCS9075 | 114.905 ms | 24 - 27 MB | NPU
75
- | Segformer-Base | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 74.118 ms | 22 - 214 MB | NPU
76
- | Segformer-Base | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 74.056 ms | 23 - 214 MB | NPU
77
- | Segformer-Base | ONNX | w8a16 | Snapdragon® X2 Elite | 6.832 ms | 17 - 17 MB | NPU
78
- | Segformer-Base | ONNX | w8a16 | Snapdragon® X Elite | 15.286 ms | 18 - 18 MB | NPU
79
- | Segformer-Base | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 10.394 ms | 14 - 249 MB | NPU
80
- | Segformer-Base | ONNX | w8a16 | Qualcomm® QCS6490 | 740.682 ms | 379 - 385 MB | CPU
81
- | Segformer-Base | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 14.768 ms | 9 - 17 MB | NPU
82
- | Segformer-Base | ONNX | w8a16 | Qualcomm® QCS9075 | 20.487 ms | 14 - 16 MB | NPU
83
- | Segformer-Base | ONNX | w8a16 | Qualcomm® QCM6690 | 352.366 ms | 332 - 342 MB | CPU
84
- | Segformer-Base | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 8.277 ms | 13 - 216 MB | NPU
85
- | Segformer-Base | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 318.353 ms | 321 - 332 MB | CPU
86
- | Segformer-Base | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 6.789 ms | 14 - 223 MB | NPU
87
- | Segformer-Base | ONNX | w8a8 | Snapdragon® X2 Elite | 4.572 ms | 3 - 3 MB | NPU
88
- | Segformer-Base | ONNX | w8a8 | Snapdragon® X Elite | 11.662 ms | 9 - 9 MB | NPU
89
- | Segformer-Base | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 7.599 ms | 7 - 235 MB | NPU
90
- | Segformer-Base | ONNX | w8a8 | Qualcomm® QCS6490 | 273.814 ms | 192 - 201 MB | CPU
91
- | Segformer-Base | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 10.971 ms | 5 - 12 MB | NPU
92
- | Segformer-Base | ONNX | w8a8 | Qualcomm® QCS9075 | 11.564 ms | 7 - 10 MB | NPU
93
- | Segformer-Base | ONNX | w8a8 | Qualcomm® QCM6690 | 174.851 ms | 193 - 205 MB | CPU
94
- | Segformer-Base | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 5.565 ms | 9 - 206 MB | NPU
95
- | Segformer-Base | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 157.155 ms | 193 - 205 MB | CPU
96
- | Segformer-Base | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 4.592 ms | 7 - 208 MB | NPU
97
- | Segformer-Base | QNN_DLC | float | Snapdragon® X2 Elite | 73.239 ms | 3 - 3 MB | NPU
98
- | Segformer-Base | QNN_DLC | float | Snapdragon® X Elite | 114.41 ms | 3 - 3 MB | NPU
99
- | Segformer-Base | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 84.1 ms | 0 - 225 MB | NPU
100
- | Segformer-Base | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 214.722 ms | 1 - 182 MB | NPU
101
- | Segformer-Base | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 110.019 ms | 3 - 144 MB | NPU
102
- | Segformer-Base | QNN_DLC | float | Qualcomm® SA8775P | 100.868 ms | 1 - 188 MB | NPU
103
- | Segformer-Base | QNN_DLC | float | Qualcomm® QCS9075 | 113.309 ms | 3 - 17 MB | NPU
104
- | Segformer-Base | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 122.016 ms | 3 - 225 MB | NPU
105
- | Segformer-Base | QNN_DLC | float | Qualcomm® SA7255P | 214.722 ms | 1 - 182 MB | NPU
106
- | Segformer-Base | QNN_DLC | float | Qualcomm® SA8295P | 122.23 ms | 1 - 184 MB | NPU
107
- | Segformer-Base | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 74.924 ms | 0 - 192 MB | NPU
108
- | Segformer-Base | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 73.878 ms | 3 - 196 MB | NPU
109
- | Segformer-Base | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 83.611 ms | 8 - 236 MB | NPU
110
- | Segformer-Base | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 214.837 ms | 0 - 183 MB | NPU
111
- | Segformer-Base | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 110.076 ms | 10 - 133 MB | NPU
112
- | Segformer-Base | TFLITE | float | Qualcomm® SA8775P | 101.022 ms | 10 - 198 MB | NPU
113
- | Segformer-Base | TFLITE | float | Qualcomm® QCS9075 | 114.489 ms | 8 - 31 MB | NPU
114
- | Segformer-Base | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 121.678 ms | 9 - 233 MB | NPU
115
- | Segformer-Base | TFLITE | float | Qualcomm® SA7255P | 214.837 ms | 0 - 183 MB | NPU
116
- | Segformer-Base | TFLITE | float | Qualcomm® SA8295P | 122.193 ms | 0 - 178 MB | NPU
117
- | Segformer-Base | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 74.961 ms | 8 - 207 MB | NPU
118
- | Segformer-Base | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 74.083 ms | 14 - 211 MB | NPU
119
- | Segformer-Base | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 10.241 ms | 2 - 210 MB | NPU
120
- | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCS6490 | 151.377 ms | 13 - 48 MB | NPU
121
- | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 23.117 ms | 2 - 175 MB | NPU
122
- | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 14.052 ms | 2 - 6 MB | NPU
123
- | Segformer-Base | TFLITE | w8a8 | Qualcomm® SA8775P | 14.434 ms | 2 - 176 MB | NPU
124
- | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCS9075 | 13.06 ms | 2 - 12 MB | NPU
125
- | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCM6690 | 152.916 ms | 13 - 176 MB | NPU
126
- | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 18.962 ms | 2 - 214 MB | NPU
127
- | Segformer-Base | TFLITE | w8a8 | Qualcomm® SA7255P | 23.117 ms | 2 - 175 MB | NPU
128
- | Segformer-Base | TFLITE | w8a8 | Qualcomm® SA8295P | 17.886 ms | 2 - 179 MB | NPU
129
- | Segformer-Base | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 7.833 ms | 2 - 175 MB | NPU
130
- | Segformer-Base | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 47.205 ms | 15 - 174 MB | NPU
131
- | Segformer-Base | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 7.2 ms | 1 - 179 MB | NPU
132
 
133
  ## License
134
  * The license for the original implementation of Segformer-Base can be found
 
15
  Segformer Base is a machine learning model that predicts masks and classes of objects in an image.
16
 
17
  This is based on the implementation of Segformer-Base found [here](https://github.com/huggingface/transformers/tree/main/src/transformers/models/segformer).
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/tree/v0.49.1/qai_hub_models/models/segformer_base) 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 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.49.1/segformer_base-onnx-float.zip)
32
+ | ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.49.1/segformer_base-onnx-w8a16.zip)
33
+ | ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.49.1/segformer_base-onnx-w8a8.zip)
34
+ | QNN_DLC | float | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.49.1/segformer_base-qnn_dlc-float.zip)
35
+ | QNN_DLC | w8a16 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.49.1/segformer_base-qnn_dlc-w8a16.zip)
36
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.43 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.49.1/segformer_base-qnn_dlc-w8a8.zip)
37
+ | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.49.1/segformer_base-tflite-float.zip)
38
+ | TFLITE | w8a8 | Universal | QAIRT 2.43, TFLite 2.17.0 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/segformer_base/releases/v0.49.1/segformer_base-tflite-w8a8.zip)
39
 
40
  For more device-specific assets and performance metrics, visit **[Segformer-Base on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/segformer_base)**.
41
 
42
 
43
  ### Option 2: Export with Custom Configurations
44
 
45
+ Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/tree/v0.49.1/qai_hub_models/models/segformer_base) Python library to compile and export the model with your own:
46
  - Custom weights (e.g., fine-tuned checkpoints)
47
  - Custom input shapes
48
  - Target device and runtime configurations
49
 
50
  This option is ideal if you need to customize the model beyond the default configuration provided here.
51
 
52
+ See our repository for [Segformer-Base on GitHub](https://github.com/qualcomm/ai-hub-models/tree/v0.49.1/qai_hub_models/models/segformer_base) for usage instructions.
53
 
54
  ## Model Details
55
 
 
67
  ## Performance Summary
68
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
69
  |---|---|---|---|---|---|---
70
+ | Segformer-Base | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 73.877 ms | 24 - 220 MB | NPU
71
+ | Segformer-Base | ONNX | float | Snapdragon® X2 Elite | 72.502 ms | 34 - 34 MB | NPU
72
+ | Segformer-Base | ONNX | float | Snapdragon® X Elite | 112.259 ms | 33 - 33 MB | NPU
73
+ | Segformer-Base | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 82.353 ms | 23 - 254 MB | NPU
74
+ | Segformer-Base | ONNX | float | Qualcomm® QCS8550 (Proxy) | 107.751 ms | 19 - 28 MB | NPU
75
+ | Segformer-Base | ONNX | float | Qualcomm® QCS9075 | 114.137 ms | 23 - 26 MB | NPU
76
+ | Segformer-Base | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 74.124 ms | 22 - 213 MB | NPU
77
+ | Segformer-Base | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 6.805 ms | 14 - 222 MB | NPU
78
+ | Segformer-Base | ONNX | w8a16 | Snapdragon® X2 Elite | 9.176 ms | 17 - 17 MB | NPU
79
+ | Segformer-Base | ONNX | w8a16 | Snapdragon® X Elite | 15.314 ms | 18 - 18 MB | NPU
80
+ | Segformer-Base | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 10.361 ms | 15 - 250 MB | NPU
81
+ | Segformer-Base | ONNX | w8a16 | Qualcomm® QCS6490 | 730.684 ms | 379 - 385 MB | CPU
82
+ | Segformer-Base | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 14.847 ms | 9 - 17 MB | NPU
83
+ | Segformer-Base | ONNX | w8a16 | Qualcomm® QCS9075 | 20.181 ms | 11 - 14 MB | NPU
84
+ | Segformer-Base | ONNX | w8a16 | Qualcomm® QCM6690 | 351.933 ms | 328 - 338 MB | CPU
85
+ | Segformer-Base | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 8.273 ms | 14 - 217 MB | NPU
86
+ | Segformer-Base | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 318.644 ms | 315 - 325 MB | CPU
87
+ | Segformer-Base | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 4.592 ms | 13 - 212 MB | NPU
88
+ | Segformer-Base | ONNX | w8a8 | Snapdragon® X2 Elite | 4.582 ms | 4 - 4 MB | NPU
89
+ | Segformer-Base | ONNX | w8a8 | Snapdragon® X Elite | 11.682 ms | 9 - 9 MB | NPU
90
+ | Segformer-Base | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 7.593 ms | 8 - 231 MB | NPU
91
+ | Segformer-Base | ONNX | w8a8 | Qualcomm® QCS6490 | 273.777 ms | 194 - 202 MB | CPU
92
+ | Segformer-Base | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 10.979 ms | 5 - 12 MB | NPU
93
+ | Segformer-Base | ONNX | w8a8 | Qualcomm® QCS9075 | 11.562 ms | 8 - 11 MB | NPU
94
+ | Segformer-Base | ONNX | w8a8 | Qualcomm® QCM6690 | 174.106 ms | 194 - 205 MB | CPU
95
+ | Segformer-Base | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 5.553 ms | 7 - 202 MB | NPU
96
+ | Segformer-Base | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 158.066 ms | 196 - 207 MB | CPU
97
+ | Segformer-Base | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 73.91 ms | 3 - 196 MB | NPU
98
+ | Segformer-Base | QNN_DLC | float | Snapdragon® X2 Elite | 73.263 ms | 3 - 3 MB | NPU
99
+ | Segformer-Base | QNN_DLC | float | Snapdragon® X Elite | 114.443 ms | 3 - 3 MB | NPU
100
+ | Segformer-Base | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 83.664 ms | 0 - 227 MB | NPU
101
+ | Segformer-Base | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 214.711 ms | 0 - 182 MB | NPU
102
+ | Segformer-Base | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 110.03 ms | 3 - 5 MB | NPU
103
+ | Segformer-Base | QNN_DLC | float | Qualcomm® SA8775P | 100.866 ms | 0 - 188 MB | NPU
104
+ | Segformer-Base | QNN_DLC | float | Qualcomm® QCS9075 | 113.333 ms | 3 - 17 MB | NPU
105
+ | Segformer-Base | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 122.051 ms | 3 - 225 MB | NPU
106
+ | Segformer-Base | QNN_DLC | float | Qualcomm® SA7255P | 214.711 ms | 0 - 182 MB | NPU
107
+ | Segformer-Base | QNN_DLC | float | Qualcomm® SA8295P | 122.162 ms | 0 - 184 MB | NPU
108
+ | Segformer-Base | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 74.95 ms | 0 - 196 MB | NPU
109
+ | Segformer-Base | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 74.245 ms | 15 - 210 MB | NPU
110
+ | Segformer-Base | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 83.311 ms | 9 - 236 MB | NPU
111
+ | Segformer-Base | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 214.83 ms | 0 - 183 MB | NPU
112
+ | Segformer-Base | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 109.794 ms | 9 - 12 MB | NPU
113
+ | Segformer-Base | TFLITE | float | Qualcomm® SA8775P | 101.021 ms | 9 - 197 MB | NPU
114
+ | Segformer-Base | TFLITE | float | Qualcomm® QCS9075 | 113.495 ms | 8 - 31 MB | NPU
115
+ | Segformer-Base | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 121.944 ms | 10 - 237 MB | NPU
116
+ | Segformer-Base | TFLITE | float | Qualcomm® SA7255P | 214.83 ms | 0 - 183 MB | NPU
117
+ | Segformer-Base | TFLITE | float | Qualcomm® SA8295P | 122.201 ms | 9 - 194 MB | NPU
118
+ | Segformer-Base | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 74.823 ms | 8 - 199 MB | NPU
119
+ | Segformer-Base | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 7.196 ms | 2 - 184 MB | NPU
120
+ | Segformer-Base | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 10.232 ms | 1 - 212 MB | NPU
121
+ | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCS6490 | 134.482 ms | 15 - 50 MB | NPU
122
+ | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 23.113 ms | 2 - 175 MB | NPU
123
+ | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 14.084 ms | 2 - 5 MB | NPU
124
+ | Segformer-Base | TFLITE | w8a8 | Qualcomm® SA8775P | 14.43 ms | 2 - 177 MB | NPU
125
+ | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCS9075 | 12.606 ms | 2 - 12 MB | NPU
126
+ | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCM6690 | 155.763 ms | 15 - 177 MB | NPU
127
+ | Segformer-Base | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 18.766 ms | 2 - 213 MB | NPU
128
+ | Segformer-Base | TFLITE | w8a8 | Qualcomm® SA7255P | 23.113 ms | 2 - 175 MB | NPU
129
+ | Segformer-Base | TFLITE | w8a8 | Qualcomm® SA8295P | 17.891 ms | 2 - 179 MB | NPU
130
+ | Segformer-Base | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 7.826 ms | 1 - 173 MB | NPU
131
+ | Segformer-Base | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 46.501 ms | 15 - 175 MB | NPU
132
 
133
  ## License
134
  * The license for the original implementation of Segformer-Base can be found