qaihm-bot commited on
Commit
325e5d5
·
verified ·
1 Parent(s): f0206ca

See https://github.com/qualcomm/ai-hub-models/releases/v0.53.0 for changelog.

Files changed (2) hide show
  1. README.md +139 -108
  2. release_assets.json +25 -25
README.md CHANGED
@@ -28,15 +28,15 @@ 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 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-onnx-float.zip)
32
- | ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-onnx-w8a16.zip)
33
- | ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-onnx-w8a8.zip)
34
- | ONNX | w8a8_mixed_int16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-onnx-w8a8_mixed_int16.zip)
35
- | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-qnn_dlc-float.zip)
36
- | QNN_DLC | w8a16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-qnn_dlc-w8a16.zip)
37
- | QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-qnn_dlc-w8a8.zip)
38
- | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-tflite-float.zip)
39
- | TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-tflite-w8a8.zip)
40
 
41
  For more device-specific assets and performance metrics, visit **[VIT on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/vit)**.
42
 
@@ -67,105 +67,136 @@ See our repository for [VIT on GitHub](https://github.com/qualcomm/ai-hub-models
67
  ## Performance Summary
68
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
69
  |---|---|---|---|---|---|---
70
- | VIT | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.662 ms | 0 - 354 MB | NPU
71
- | VIT | ONNX | float | Snapdragon® X2 Elite | 3.873 ms | 170 - 170 MB | NPU
72
- | VIT | ONNX | float | Snapdragon® X Elite | 11.128 ms | 170 - 170 MB | NPU
73
- | VIT | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 7.171 ms | 0 - 377 MB | NPU
74
- | VIT | ONNX | float | Qualcomm® QCS8550 (Proxy) | 10.454 ms | 0 - 195 MB | NPU
75
- | VIT | ONNX | float | Qualcomm® QCS9075 | 14.697 ms | 0 - 4 MB | NPU
76
- | VIT | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.99 ms | 0 - 348 MB | NPU
77
- | VIT | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 3.775 ms | 0 - 303 MB | NPU
78
- | VIT | ONNX | w8a16 | Snapdragon® X2 Elite | 3.902 ms | 86 - 86 MB | NPU
79
- | VIT | ONNX | w8a16 | Snapdragon® X Elite | 11.253 ms | 86 - 86 MB | NPU
80
- | VIT | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 7.363 ms | 0 - 378 MB | NPU
81
- | VIT | ONNX | w8a16 | Qualcomm® QCS6490 | 1117.068 ms | 38 - 59 MB | CPU
82
- | VIT | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 10.668 ms | 0 - 101 MB | NPU
83
- | VIT | ONNX | w8a16 | Qualcomm® QCS9075 | 12.939 ms | 0 - 3 MB | NPU
84
- | VIT | ONNX | w8a16 | Qualcomm® QCM6690 | 611.108 ms | 73 - 88 MB | CPU
85
- | VIT | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 5.314 ms | 0 - 300 MB | NPU
86
- | VIT | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 593.078 ms | 60 - 76 MB | CPU
87
- | VIT | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 4.582 ms | 0 - 351 MB | NPU
88
- | VIT | ONNX | w8a8 | Snapdragon® X2 Elite | 5.101 ms | 85 - 85 MB | NPU
89
- | VIT | ONNX | w8a8 | Snapdragon® X Elite | 13.62 ms | 85 - 85 MB | NPU
90
- | VIT | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 8.792 ms | 0 - 457 MB | NPU
91
- | VIT | ONNX | w8a8 | Qualcomm® QCS6490 | 316.911 ms | 21 - 74 MB | CPU
92
- | VIT | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 12.93 ms | 0 - 100 MB | NPU
93
- | VIT | ONNX | w8a8 | Qualcomm® QCS9075 | 13.549 ms | 0 - 3 MB | NPU
94
- | VIT | ONNX | w8a8 | Qualcomm® QCM6690 | 136.498 ms | 21 - 42 MB | CPU
95
- | VIT | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 7.243 ms | 0 - 323 MB | NPU
96
- | VIT | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 129.249 ms | 22 - 43 MB | CPU
97
- | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 62.922 ms | 21 - 289 MB | NPU
98
- | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® X2 Elite | 56.849 ms | 79 - 79 MB | NPU
99
- | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® X Elite | 160.029 ms | 79 - 79 MB | NPU
100
- | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 85.742 ms | 69 - 445 MB | NPU
101
- | VIT | ONNX | w8a8_mixed_int16 | Qualcomm® QCS6490 | 733.594 ms | 96 - 128 MB | CPU
102
- | VIT | ONNX | w8a8_mixed_int16 | Qualcomm® QCS8550 (Proxy) | 100.182 ms | 9 - 95 MB | NPU
103
- | VIT | ONNX | w8a8_mixed_int16 | Qualcomm® QCS9075 | 130.409 ms | 68 - 70 MB | NPU
104
- | VIT | ONNX | w8a8_mixed_int16 | Qualcomm® QCM6690 | 383.515 ms | 99 - 121 MB | CPU
105
- | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Elite For Galaxy Mobile | 72.88 ms | 68 - 339 MB | NPU
106
- | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 373.213 ms | 95 - 114 MB | CPU
107
- | VIT | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 4.099 ms | 1 - 231 MB | NPU
108
- | VIT | QNN_DLC | float | Snapdragon® X2 Elite | 4.492 ms | 1 - 1 MB | NPU
109
- | VIT | QNN_DLC | float | Snapdragon® X Elite | 11.9 ms | 1 - 1 MB | NPU
110
- | VIT | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 7.683 ms | 0 - 309 MB | NPU
111
- | VIT | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 40.628 ms | 1 - 177 MB | NPU
112
- | VIT | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 11.182 ms | 1 - 2 MB | NPU
113
- | VIT | QNN_DLC | float | Qualcomm® SA8775P | 13.985 ms | 1 - 213 MB | NPU
114
- | VIT | QNN_DLC | float | Qualcomm® QCS9075 | 15.354 ms | 1 - 3 MB | NPU
115
- | VIT | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 19.231 ms | 0 - 290 MB | NPU
116
- | VIT | QNN_DLC | float | Qualcomm® SA7255P | 40.628 ms | 1 - 177 MB | NPU
117
- | VIT | QNN_DLC | float | Qualcomm® SA8295P | 17.266 ms | 0 - 193 MB | NPU
118
- | VIT | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 5.353 ms | 1 - 224 MB | NPU
119
- | VIT | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 3.378 ms | 0 - 299 MB | NPU
120
- | VIT | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 4.176 ms | 0 - 0 MB | NPU
121
- | VIT | QNN_DLC | w8a16 | Snapdragon® X Elite | 9.875 ms | 0 - 0 MB | NPU
122
- | VIT | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 6.454 ms | 0 - 345 MB | NPU
123
- | VIT | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 20.263 ms | 0 - 283 MB | NPU
124
- | VIT | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 9.164 ms | 0 - 305 MB | NPU
125
- | VIT | QNN_DLC | w8a16 | Qualcomm® SA8775P | 9.406 ms | 0 - 285 MB | NPU
126
- | VIT | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 11.278 ms | 0 - 2 MB | NPU
127
- | VIT | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 87.105 ms | 0 - 413 MB | NPU
128
- | VIT | QNN_DLC | w8a16 | Qualcomm® SA7255P | 20.263 ms | 0 - 283 MB | NPU
129
- | VIT | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 4.731 ms | 0 - 285 MB | NPU
130
- | VIT | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 15.048 ms | 0 - 430 MB | NPU
131
- | VIT | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 3.705 ms | 0 - 280 MB | NPU
132
- | VIT | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 4.607 ms | 0 - 0 MB | NPU
133
- | VIT | QNN_DLC | w8a8 | Snapdragon® X Elite | 10.736 ms | 0 - 0 MB | NPU
134
- | VIT | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 7.075 ms | 0 - 410 MB | NPU
135
- | VIT | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 50.746 ms | 2 - 4 MB | NPU
136
- | VIT | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 29.678 ms | 0 - 288 MB | NPU
137
- | VIT | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 10.224 ms | 0 - 2 MB | NPU
138
- | VIT | QNN_DLC | w8a8 | Qualcomm® SA8775P | 9.331 ms | 0 - 288 MB | NPU
139
- | VIT | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 9.905 ms | 0 - 2 MB | NPU
140
- | VIT | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 217.898 ms | 0 - 502 MB | NPU
141
- | VIT | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 13.23 ms | 0 - 418 MB | NPU
142
- | VIT | QNN_DLC | w8a8 | Qualcomm® SA7255P | 29.678 ms | 0 - 288 MB | NPU
143
- | VIT | QNN_DLC | w8a8 | Qualcomm® SA8295P | 16.102 ms | 0 - 188 MB | NPU
144
- | VIT | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 5.848 ms | 0 - 288 MB | NPU
145
- | VIT | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 20.003 ms | 0 - 327 MB | NPU
146
- | VIT | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.029 ms | 0 - 174 MB | NPU
147
- | VIT | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 5.698 ms | 0 - 274 MB | NPU
148
- | VIT | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 35.349 ms | 0 - 187 MB | NPU
149
- | VIT | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 7.866 ms | 0 - 3 MB | NPU
150
- | VIT | TFLITE | float | Qualcomm® SA8775P | 10.791 ms | 0 - 190 MB | NPU
151
- | VIT | TFLITE | float | Qualcomm® QCS9075 | 11.475 ms | 0 - 173 MB | NPU
152
- | VIT | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 13.903 ms | 0 - 241 MB | NPU
153
- | VIT | TFLITE | float | Qualcomm® SA7255P | 35.349 ms | 0 - 187 MB | NPU
154
- | VIT | TFLITE | float | Qualcomm® SA8295P | 13.431 ms | 0 - 154 MB | NPU
155
- | VIT | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.908 ms | 0 - 184 MB | NPU
156
- | VIT | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 4.569 ms | 0 - 334 MB | NPU
157
- | VIT | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 8.776 ms | 0 - 457 MB | NPU
158
- | VIT | TFLITE | w8a8 | Qualcomm® QCS6490 | 152.222 ms | 1 - 101 MB | NPU
159
- | VIT | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 35.43 ms | 0 - 327 MB | NPU
160
- | VIT | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 12.279 ms | 0 - 4 MB | NPU
161
- | VIT | TFLITE | w8a8 | Qualcomm® SA8775P | 12.377 ms | 0 - 327 MB | NPU
162
- | VIT | TFLITE | w8a8 | Qualcomm® QCS9075 | 13.843 ms | 0 - 88 MB | NPU
163
- | VIT | TFLITE | w8a8 | Qualcomm® QCM6690 | 203.01 ms | 2 - 287 MB | NPU
164
- | VIT | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 21.722 ms | 0 - 446 MB | NPU
165
- | VIT | TFLITE | w8a8 | Qualcomm® SA7255P | 35.43 ms | 0 - 327 MB | NPU
166
- | VIT | TFLITE | w8a8 | Qualcomm® SA8295P | 19.859 ms | 0 - 318 MB | NPU
167
- | VIT | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 7.09 ms | 0 - 317 MB | NPU
168
- | VIT | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 30.808 ms | 2 - 191 MB | NPU
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
169
 
170
  ## License
171
  * The license for the original implementation of VIT can be found
 
28
 
29
  | Runtime | Precision | Chipset | SDK Versions | Download |
30
  |---|---|---|---|---|
31
+ | ONNX | float | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-onnx-float.zip)
32
+ | ONNX | w8a16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-onnx-w8a16.zip)
33
+ | ONNX | w8a8 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-onnx-w8a8.zip)
34
+ | ONNX | w8a8_mixed_int16 | Universal | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-onnx-w8a8_mixed_int16.zip)
35
+ | QNN_DLC | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-qnn_dlc-float.zip)
36
+ | QNN_DLC | w8a16 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-qnn_dlc-w8a16.zip)
37
+ | QNN_DLC | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-qnn_dlc-w8a8.zip)
38
+ | TFLITE | float | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-tflite-float.zip)
39
+ | TFLITE | w8a8 | Universal | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-tflite-w8a8.zip)
40
 
41
  For more device-specific assets and performance metrics, visit **[VIT on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/vit)**.
42
 
 
67
  ## Performance Summary
68
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
69
  |---|---|---|---|---|---|---
70
+ | VIT | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.697 ms | 1 - 346 MB | NPU
71
+ | VIT | ONNX | float | Snapdragon® 8 Elite Mobile | 3.445 ms | 0 - 336 MB | NPU
72
+ | VIT | ONNX | float | Snapdragon® X2 Elite | 3.04 ms | 170 - 170 MB | NPU
73
+ | VIT | ONNX | float | Snapdragon® X Elite | 7.674 ms | 170 - 170 MB | NPU
74
+ | VIT | ONNX | float | Snapdragon® X Elite | 7.674 ms | 170 - 170 MB | NPU
75
+ | VIT | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 5.034 ms | 0 - 366 MB | NPU
76
+ | VIT | ONNX | float | Qualcomm® QCS8550 (Proxy) | 7.158 ms | 0 - 195 MB | NPU
77
+ | VIT | ONNX | float | Qualcomm® QCS9075 | 10.156 ms | 0 - 4 MB | NPU
78
+ | VIT | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.445 ms | 0 - 336 MB | NPU
79
+ | VIT | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 3.123 ms | 0 - 274 MB | NPU
80
+ | VIT | ONNX | w8a16 | Snapdragon® 8 Elite Mobile | 4.408 ms | 0 - 269 MB | NPU
81
+ | VIT | ONNX | w8a16 | Snapdragon® X2 Elite | 3.182 ms | 86 - 86 MB | NPU
82
+ | VIT | ONNX | w8a16 | Snapdragon® X Elite | 8.804 ms | 86 - 86 MB | NPU
83
+ | VIT | ONNX | w8a16 | Snapdragon® X Elite | 8.804 ms | 86 - 86 MB | NPU
84
+ | VIT | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 5.697 ms | 0 - 359 MB | NPU
85
+ | VIT | ONNX | w8a16 | Qualcomm® QCS6490 | 1104.029 ms | 43 - 59 MB | CPU
86
+ | VIT | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 8.17 ms | 0 - 5 MB | NPU
87
+ | VIT | ONNX | w8a16 | Qualcomm® QCS9075 | 8.791 ms | 0 - 3 MB | NPU
88
+ | VIT | ONNX | w8a16 | Qualcomm® QCM6690 | 605.462 ms | 75 - 90 MB | CPU
89
+ | VIT | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 4.408 ms | 0 - 269 MB | NPU
90
+ | VIT | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 591.548 ms | 60 - 76 MB | CPU
91
+ | VIT | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 591.548 ms | 60 - 76 MB | CPU
92
+ | VIT | ONNX | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 4.578 ms | 0 - 420 MB | NPU
93
+ | VIT | ONNX | w8a8 | Snapdragon® 8 Elite Mobile | 6.956 ms | 0 - 415 MB | NPU
94
+ | VIT | ONNX | w8a8 | Snapdragon® X2 Elite | 5.211 ms | 85 - 85 MB | NPU
95
+ | VIT | ONNX | w8a8 | Snapdragon® X Elite | 13.687 ms | 85 - 85 MB | NPU
96
+ | VIT | ONNX | w8a8 | Snapdragon® X Elite | 13.687 ms | 85 - 85 MB | NPU
97
+ | VIT | ONNX | w8a8 | Snapdragon® 8 Gen 3 Mobile | 8.854 ms | 0 - 522 MB | NPU
98
+ | VIT | ONNX | w8a8 | Qualcomm® QCS6490 | 319.536 ms | 20 - 70 MB | CPU
99
+ | VIT | ONNX | w8a8 | Qualcomm® QCS8550 (Proxy) | 13.028 ms | 0 - 104 MB | NPU
100
+ | VIT | ONNX | w8a8 | Qualcomm® QCS9075 | 14.087 ms | 0 - 3 MB | NPU
101
+ | VIT | ONNX | w8a8 | Qualcomm® QCM6690 | 134.03 ms | 21 - 44 MB | CPU
102
+ | VIT | ONNX | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 6.956 ms | 0 - 415 MB | NPU
103
+ | VIT | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 126.126 ms | 24 - 44 MB | CPU
104
+ | VIT | ONNX | w8a8 | Snapdragon® 7 Gen 4 Mobile | 126.126 ms | 24 - 44 MB | CPU
105
+ | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Elite Gen 5 Mobile | 72.821 ms | 77 - 386 MB | NPU
106
+ | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Elite Mobile | 84.782 ms | 72 - 369 MB | NPU
107
+ | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® X2 Elite | 57.918 ms | 137 - 137 MB | NPU
108
+ | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® X Elite | 267.139 ms | 136 - 136 MB | NPU
109
+ | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® X Elite | 267.139 ms | 136 - 136 MB | NPU
110
+ | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Gen 3 Mobile | 99.614 ms | 78 - 465 MB | NPU
111
+ | VIT | ONNX | w8a8_mixed_int16 | Qualcomm® QCS6490 | 752.275 ms | 98 - 127 MB | CPU
112
+ | VIT | ONNX | w8a8_mixed_int16 | Qualcomm® QCS9075 | 137.424 ms | 77 - 80 MB | NPU
113
+ | VIT | ONNX | w8a8_mixed_int16 | Qualcomm® QCM6690 | 393.279 ms | 104 - 125 MB | CPU
114
+ | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® 8 Elite For Galaxy Mobile | 84.782 ms | 72 - 369 MB | NPU
115
+ | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 383.282 ms | 92 - 111 MB | CPU
116
+ | VIT | ONNX | w8a8_mixed_int16 | Snapdragon® 7 Gen 4 Mobile | 383.282 ms | 92 - 111 MB | CPU
117
+ | VIT | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.918 ms | 1 - 155 MB | NPU
118
+ | VIT | QNN_DLC | float | Snapdragon® 8 Elite Mobile | 3.726 ms | 1 - 159 MB | NPU
119
+ | VIT | QNN_DLC | float | Snapdragon® X2 Elite | 3.504 ms | 1 - 1 MB | NPU
120
+ | VIT | QNN_DLC | float | Snapdragon® X Elite | 8.276 ms | 1 - 1 MB | NPU
121
+ | VIT | QNN_DLC | float | Snapdragon® X Elite | 8.276 ms | 1 - 1 MB | NPU
122
+ | VIT | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 5.394 ms | 0 - 252 MB | NPU
123
+ | VIT | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 34.775 ms | 1 - 168 MB | NPU
124
+ | VIT | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 7.708 ms | 1 - 398 MB | NPU
125
+ | VIT | QNN_DLC | float | Qualcomm® SA8775P | 10.456 ms | 1 - 168 MB | NPU
126
+ | VIT | QNN_DLC | float | Qualcomm® SA8775P | 10.456 ms | 1 - 168 MB | NPU
127
+ | VIT | QNN_DLC | float | Qualcomm® SA8775P | 10.456 ms | 1 - 168 MB | NPU
128
+ | VIT | QNN_DLC | float | Qualcomm® QCS9075 | 10.685 ms | 1 - 3 MB | NPU
129
+ | VIT | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 13.207 ms | 0 - 229 MB | NPU
130
+ | VIT | QNN_DLC | float | Qualcomm® SA7255P | 34.775 ms | 1 - 168 MB | NPU
131
+ | VIT | QNN_DLC | float | Qualcomm® SA8295P | 12.957 ms | 1 - 145 MB | NPU
132
+ | VIT | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.726 ms | 1 - 159 MB | NPU
133
+ | VIT | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 2.82 ms | 0 - 274 MB | NPU
134
+ | VIT | QNN_DLC | w8a16 | Snapdragon® 8 Elite Mobile | 3.902 ms | 0 - 258 MB | NPU
135
+ | VIT | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 3.593 ms | 0 - 0 MB | NPU
136
+ | VIT | QNN_DLC | w8a16 | Snapdragon® X Elite | 8.224 ms | 0 - 0 MB | NPU
137
+ | VIT | QNN_DLC | w8a16 | Snapdragon® X Elite | 8.224 ms | 0 - 0 MB | NPU
138
+ | VIT | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 5.085 ms | 0 - 321 MB | NPU
139
+ | VIT | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 17.365 ms | 0 - 257 MB | NPU
140
+ | VIT | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 7.645 ms | 0 - 2 MB | NPU
141
+ | VIT | QNN_DLC | w8a16 | Qualcomm® SA8775P | 7.793 ms | 0 - 258 MB | NPU
142
+ | VIT | QNN_DLC | w8a16 | Qualcomm® SA8775P | 7.793 ms | 0 - 258 MB | NPU
143
+ | VIT | QNN_DLC | w8a16 | Qualcomm® SA8775P | 7.793 ms | 0 - 258 MB | NPU
144
+ | VIT | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 7.981 ms | 0 - 2 MB | NPU
145
+ | VIT | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 86.803 ms | 0 - 404 MB | NPU
146
+ | VIT | QNN_DLC | w8a16 | Qualcomm® SA7255P | 17.365 ms | 0 - 257 MB | NPU
147
+ | VIT | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 3.902 ms | 0 - 258 MB | NPU
148
+ | VIT | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 12.919 ms | 0 - 402 MB | NPU
149
+ | VIT | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 12.919 ms | 0 - 402 MB | NPU
150
+ | VIT | QNN_DLC | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 3.495 ms | 0 - 207 MB | NPU
151
+ | VIT | QNN_DLC | w8a8 | Snapdragon® 8 Elite Mobile | 5.415 ms | 52 - 253 MB | NPU
152
+ | VIT | QNN_DLC | w8a8 | Snapdragon® X2 Elite | 4.488 ms | 0 - 0 MB | NPU
153
+ | VIT | QNN_DLC | w8a8 | Snapdragon® X Elite | 10.193 ms | 0 - 0 MB | NPU
154
+ | VIT | QNN_DLC | w8a8 | Snapdragon® X Elite | 10.193 ms | 0 - 0 MB | NPU
155
+ | VIT | QNN_DLC | w8a8 | Snapdragon® 8 Gen 3 Mobile | 6.472 ms | 0 - 305 MB | NPU
156
+ | VIT | QNN_DLC | w8a8 | Qualcomm® QCS6490 | 50.546 ms | 0 - 2 MB | NPU
157
+ | VIT | QNN_DLC | w8a8 | Qualcomm® QCS8275 (Proxy) | 28.466 ms | 0 - 200 MB | NPU
158
+ | VIT | QNN_DLC | w8a8 | Qualcomm® QCS8550 (Proxy) | 9.567 ms | 0 - 2 MB | NPU
159
+ | VIT | QNN_DLC | w8a8 | Qualcomm® SA8775P | 8.659 ms | 0 - 201 MB | NPU
160
+ | VIT | QNN_DLC | w8a8 | Qualcomm® SA8775P | 8.659 ms | 0 - 201 MB | NPU
161
+ | VIT | QNN_DLC | w8a8 | Qualcomm® SA8775P | 8.659 ms | 0 - 201 MB | NPU
162
+ | VIT | QNN_DLC | w8a8 | Qualcomm® QCS9075 | 9.497 ms | 0 - 2 MB | NPU
163
+ | VIT | QNN_DLC | w8a8 | Qualcomm® QCM6690 | 223.386 ms | 0 - 499 MB | NPU
164
+ | VIT | QNN_DLC | w8a8 | Qualcomm® QCS8450 (Proxy) | 13.179 ms | 0 - 310 MB | NPU
165
+ | VIT | QNN_DLC | w8a8 | Qualcomm® SA7255P | 28.466 ms | 0 - 200 MB | NPU
166
+ | VIT | QNN_DLC | w8a8 | Qualcomm® SA8295P | 15.592 ms | 0 - 211 MB | NPU
167
+ | VIT | QNN_DLC | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 5.415 ms | 52 - 253 MB | NPU
168
+ | VIT | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 19.485 ms | 0 - 355 MB | NPU
169
+ | VIT | QNN_DLC | w8a8 | Snapdragon® 7 Gen 4 Mobile | 19.485 ms | 0 - 355 MB | NPU
170
+ | VIT | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.026 ms | 0 - 169 MB | NPU
171
+ | VIT | TFLITE | float | Snapdragon® 8 Elite Mobile | 3.69 ms | 0 - 175 MB | NPU
172
+ | VIT | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 5.266 ms | 0 - 303 MB | NPU
173
+ | VIT | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 34.6 ms | 0 - 169 MB | NPU
174
+ | VIT | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 7.228 ms | 0 - 4 MB | NPU
175
+ | VIT | TFLITE | float | Qualcomm® SA8775P | 10.149 ms | 0 - 185 MB | NPU
176
+ | VIT | TFLITE | float | Qualcomm® SA8775P | 10.149 ms | 0 - 185 MB | NPU
177
+ | VIT | TFLITE | float | Qualcomm® SA8775P | 10.149 ms | 0 - 185 MB | NPU
178
+ | VIT | TFLITE | float | Qualcomm® QCS9075 | 10.603 ms | 0 - 173 MB | NPU
179
+ | VIT | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 12.966 ms | 0 - 290 MB | NPU
180
+ | VIT | TFLITE | float | Qualcomm® SA7255P | 34.6 ms | 0 - 169 MB | NPU
181
+ | VIT | TFLITE | float | Qualcomm® SA8295P | 13.265 ms | 0 - 156 MB | NPU
182
+ | VIT | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 3.69 ms | 0 - 175 MB | NPU
183
+ | VIT | TFLITE | w8a8 | Snapdragon® 8 Elite Gen 5 Mobile | 4.544 ms | 0 - 379 MB | NPU
184
+ | VIT | TFLITE | w8a8 | Snapdragon® 8 Elite Mobile | 6.927 ms | 0 - 380 MB | NPU
185
+ | VIT | TFLITE | w8a8 | Snapdragon® 8 Gen 3 Mobile | 8.67 ms | 0 - 486 MB | NPU
186
+ | VIT | TFLITE | w8a8 | Qualcomm® QCS6490 | 147.125 ms | 1 - 100 MB | NPU
187
+ | VIT | TFLITE | w8a8 | Qualcomm® QCS8275 (Proxy) | 35.052 ms | 0 - 385 MB | NPU
188
+ | VIT | TFLITE | w8a8 | Qualcomm® QCS8550 (Proxy) | 12.928 ms | 0 - 3 MB | NPU
189
+ | VIT | TFLITE | w8a8 | Qualcomm® SA8775P | 12.173 ms | 0 - 387 MB | NPU
190
+ | VIT | TFLITE | w8a8 | Qualcomm® SA8775P | 12.173 ms | 0 - 387 MB | NPU
191
+ | VIT | TFLITE | w8a8 | Qualcomm® SA8775P | 12.173 ms | 0 - 387 MB | NPU
192
+ | VIT | TFLITE | w8a8 | Qualcomm® QCS9075 | 13.71 ms | 0 - 88 MB | NPU
193
+ | VIT | TFLITE | w8a8 | Qualcomm® QCM6690 | 202.096 ms | 2 - 284 MB | NPU
194
+ | VIT | TFLITE | w8a8 | Qualcomm® QCS8450 (Proxy) | 20.594 ms | 0 - 447 MB | NPU
195
+ | VIT | TFLITE | w8a8 | Qualcomm® SA7255P | 35.052 ms | 0 - 385 MB | NPU
196
+ | VIT | TFLITE | w8a8 | Qualcomm® SA8295P | 19.247 ms | 0 - 347 MB | NPU
197
+ | VIT | TFLITE | w8a8 | Snapdragon® 8 Elite For Galaxy Mobile | 6.927 ms | 0 - 380 MB | NPU
198
+ | VIT | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 29.672 ms | 2 - 215 MB | NPU
199
+ | VIT | TFLITE | w8a8 | Snapdragon® 7 Gen 4 Mobile | 29.672 ms | 2 - 215 MB | NPU
200
 
201
  ## License
202
  * The license for the original implementation of VIT can be found
release_assets.json CHANGED
@@ -1,79 +1,79 @@
1
  {
2
- "version": "0.52.0",
3
  "precisions": {
4
  "w8a8": {
5
  "universal_assets": {
6
  "tflite": {
7
  "tool_versions": {
8
  "qairt": "2.45.0.260326154327",
9
- "litert": "1.4.2"
10
  },
11
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-tflite-w8a8.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/vit/releases/v0.52.0/vit-qnn_dlc-w8a8.zip"
18
  },
19
  "onnx": {
20
  "tool_versions": {
21
  "qairt": "2.42.0.251225135753_193295",
22
  "onnx_runtime": "1.24.3"
23
  },
24
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-onnx-w8a8.zip"
25
  }
26
  }
27
  },
28
- "float": {
29
  "universal_assets": {
30
- "tflite": {
31
- "tool_versions": {
32
- "qairt": "2.45.0.260326154327",
33
- "litert": "1.4.2"
34
- },
35
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-tflite-float.zip"
36
- },
37
  "qnn_dlc": {
38
  "tool_versions": {
39
  "qairt": "2.45.0.260326154327"
40
  },
41
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-qnn_dlc-float.zip"
42
  },
43
  "onnx": {
44
  "tool_versions": {
45
  "qairt": "2.42.0.251225135753_193295",
46
  "onnx_runtime": "1.24.3"
47
  },
48
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-onnx-float.zip"
49
  }
50
  }
51
  },
52
- "w8a16": {
53
  "universal_assets": {
54
- "qnn_dlc": {
55
- "tool_versions": {
56
- "qairt": "2.45.0.260326154327"
57
- },
58
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-qnn_dlc-w8a16.zip"
59
- },
60
  "onnx": {
61
  "tool_versions": {
62
  "qairt": "2.42.0.251225135753_193295",
63
  "onnx_runtime": "1.24.3"
64
  },
65
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-onnx-w8a16.zip"
66
  }
67
  }
68
  },
69
- "w8a8_mixed_int16": {
70
  "universal_assets": {
 
 
 
 
 
 
 
 
 
 
 
 
 
71
  "onnx": {
72
  "tool_versions": {
73
  "qairt": "2.42.0.251225135753_193295",
74
  "onnx_runtime": "1.24.3"
75
  },
76
- "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.52.0/vit-onnx-w8a8_mixed_int16.zip"
77
  }
78
  }
79
  }
 
1
  {
2
+ "version": "0.53.0",
3
  "precisions": {
4
  "w8a8": {
5
  "universal_assets": {
6
  "tflite": {
7
  "tool_versions": {
8
  "qairt": "2.45.0.260326154327",
9
+ "litert": "1.4.3"
10
  },
11
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-tflite-w8a8.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/vit/releases/v0.53.0/vit-qnn_dlc-w8a8.zip"
18
  },
19
  "onnx": {
20
  "tool_versions": {
21
  "qairt": "2.42.0.251225135753_193295",
22
  "onnx_runtime": "1.24.3"
23
  },
24
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-onnx-w8a8.zip"
25
  }
26
  }
27
  },
28
+ "w8a16": {
29
  "universal_assets": {
 
 
 
 
 
 
 
30
  "qnn_dlc": {
31
  "tool_versions": {
32
  "qairt": "2.45.0.260326154327"
33
  },
34
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-qnn_dlc-w8a16.zip"
35
  },
36
  "onnx": {
37
  "tool_versions": {
38
  "qairt": "2.42.0.251225135753_193295",
39
  "onnx_runtime": "1.24.3"
40
  },
41
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-onnx-w8a16.zip"
42
  }
43
  }
44
  },
45
+ "w8a8_mixed_int16": {
46
  "universal_assets": {
 
 
 
 
 
 
47
  "onnx": {
48
  "tool_versions": {
49
  "qairt": "2.42.0.251225135753_193295",
50
  "onnx_runtime": "1.24.3"
51
  },
52
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-onnx-w8a8_mixed_int16.zip"
53
  }
54
  }
55
  },
56
+ "float": {
57
  "universal_assets": {
58
+ "tflite": {
59
+ "tool_versions": {
60
+ "qairt": "2.45.0.260326154327",
61
+ "litert": "1.4.3"
62
+ },
63
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-tflite-float.zip"
64
+ },
65
+ "qnn_dlc": {
66
+ "tool_versions": {
67
+ "qairt": "2.45.0.260326154327"
68
+ },
69
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-qnn_dlc-float.zip"
70
+ },
71
  "onnx": {
72
  "tool_versions": {
73
  "qairt": "2.42.0.251225135753_193295",
74
  "onnx_runtime": "1.24.3"
75
  },
76
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/vit/releases/v0.53.0/vit-onnx-float.zip"
77
  }
78
  }
79
  }