Image Classification
PyTorch
backbone
android
qaihm-bot commited on
Commit
059f789
·
verified ·
1 Parent(s): ed85b3e

See https://github.com/qualcomm/ai-hub-models/releases/v0.51.0 for changelog.

Files changed (2) hide show
  1. README.md +56 -30
  2. release_assets.json +46 -1
README.md CHANGED
@@ -28,9 +28,11 @@ 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/beit/releases/v0.50.2/beit-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/beit/releases/v0.50.2/beit-onnx-w8a16.zip)
33
- | TFLITE | float | Universal | QAIRT 2.43, TFLite 2.19.1 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/beit/releases/v0.50.2/beit-tflite-float.zip)
 
 
34
 
35
  For more device-specific assets and performance metrics, visit **[Beit on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/beit)**.
36
 
@@ -59,33 +61,57 @@ See our repository for [Beit on GitHub](https://github.com/qualcomm/ai-hub-model
59
  ## Performance Summary
60
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
61
  |---|---|---|---|---|---|---
62
- | Beit | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.739 ms | 1 - 476 MB | NPU
63
- | Beit | ONNX | float | Snapdragon® X2 Elite | 4.056 ms | 185 - 185 MB | NPU
64
- | Beit | ONNX | float | Snapdragon® X Elite | 9.973 ms | 185 - 185 MB | NPU
65
- | Beit | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 6.282 ms | 0 - 408 MB | NPU
66
- | Beit | ONNX | float | Qualcomm® QCS8550 (Proxy) | 9.242 ms | 0 - 196 MB | NPU
67
- | Beit | ONNX | float | Qualcomm® QCS9075 | 12.767 ms | 0 - 4 MB | NPU
68
- | Beit | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.622 ms | 0 - 483 MB | NPU
69
- | Beit | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 3.772 ms | 0 - 291 MB | NPU
70
- | Beit | ONNX | w8a16 | Snapdragon® X2 Elite | 4.104 ms | 102 - 102 MB | NPU
71
- | Beit | ONNX | w8a16 | Snapdragon® X Elite | 10.979 ms | 102 - 102 MB | NPU
72
- | Beit | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 6.965 ms | 0 - 409 MB | NPU
73
- | Beit | ONNX | w8a16 | Qualcomm® QCS6490 | 1072.127 ms | 39 - 337 MB | CPU
74
- | Beit | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 10.3 ms | 0 - 123 MB | NPU
75
- | Beit | ONNX | w8a16 | Qualcomm® QCS9075 | 10.654 ms | 0 - 3 MB | NPU
76
- | Beit | ONNX | w8a16 | Qualcomm® QCM6690 | 609.294 ms | 96 - 111 MB | CPU
77
- | Beit | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 5.734 ms | 0 - 401 MB | NPU
78
- | Beit | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 583.008 ms | 99 - 115 MB | CPU
79
- | Beit | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.373 ms | 0 - 465 MB | NPU
80
- | Beit | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 5.709 ms | 0 - 518 MB | NPU
81
- | Beit | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 36.48 ms | 0 - 471 MB | NPU
82
- | Beit | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 7.672 ms | 0 - 3 MB | NPU
83
- | Beit | TFLITE | float | Qualcomm® SA8775P | 11.083 ms | 0 - 471 MB | NPU
84
- | Beit | TFLITE | float | Qualcomm® QCS9075 | 11.696 ms | 0 - 186 MB | NPU
85
- | Beit | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 17.806 ms | 0 - 475 MB | NPU
86
- | Beit | TFLITE | float | Qualcomm® SA7255P | 36.48 ms | 0 - 471 MB | NPU
87
- | Beit | TFLITE | float | Qualcomm® SA8295P | 14.51 ms | 1 - 455 MB | NPU
88
- | Beit | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.14 ms | 0 - 477 MB | NPU
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
89
 
90
  ## License
91
  * The license for the original implementation of Beit 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/beit/releases/v0.51.0/beit-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/beit/releases/v0.51.0/beit-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/beit/releases/v0.51.0/beit-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/beit/releases/v0.51.0/beit-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/beit/releases/v0.51.0/beit-tflite-float.zip)
36
 
37
  For more device-specific assets and performance metrics, visit **[Beit on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/beit)**.
38
 
 
61
  ## Performance Summary
62
  | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit
63
  |---|---|---|---|---|---|---
64
+ | Beit | ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 6.603 ms | 0 - 493 MB | NPU
65
+ | Beit | ONNX | float | Snapdragon® X2 Elite | 6.192 ms | 185 - 185 MB | NPU
66
+ | Beit | ONNX | float | Snapdragon® X Elite | 15.501 ms | 184 - 184 MB | NPU
67
+ | Beit | ONNX | float | Snapdragon® 8 Gen 3 Mobile | 10.279 ms | 1 - 576 MB | NPU
68
+ | Beit | ONNX | float | Qualcomm® QCS8550 (Proxy) | 14.796 ms | 0 - 195 MB | NPU
69
+ | Beit | ONNX | float | Qualcomm® QCS9075 | 18.639 ms | 0 - 4 MB | NPU
70
+ | Beit | ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 8.104 ms | 1 - 509 MB | NPU
71
+ | Beit | ONNX | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 4.467 ms | 0 - 417 MB | NPU
72
+ | Beit | ONNX | w8a16 | Snapdragon® X2 Elite | 4.793 ms | 102 - 102 MB | NPU
73
+ | Beit | ONNX | w8a16 | Snapdragon® X Elite | 12.345 ms | 102 - 102 MB | NPU
74
+ | Beit | ONNX | w8a16 | Snapdragon® 8 Gen 3 Mobile | 7.861 ms | 0 - 560 MB | NPU
75
+ | Beit | ONNX | w8a16 | Qualcomm® QCS6490 | 1105.992 ms | 50 - 70 MB | CPU
76
+ | Beit | ONNX | w8a16 | Qualcomm® QCS8550 (Proxy) | 11.616 ms | 0 - 6 MB | NPU
77
+ | Beit | ONNX | w8a16 | Qualcomm® QCS9075 | 13.394 ms | 0 - 3 MB | NPU
78
+ | Beit | ONNX | w8a16 | Qualcomm® QCM6690 | 602.324 ms | 82 - 94 MB | CPU
79
+ | Beit | ONNX | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 6.464 ms | 0 - 411 MB | NPU
80
+ | Beit | ONNX | w8a16 | Snapdragon® 7 Gen 4 Mobile | 588.011 ms | 74 - 87 MB | CPU
81
+ | Beit | QNN_DLC | float | Snapdragon® 8 Elite Gen 5 Mobile | 7.624 ms | 1 - 358 MB | NPU
82
+ | Beit | QNN_DLC | float | Snapdragon® X2 Elite | 7.813 ms | 1 - 1 MB | NPU
83
+ | Beit | QNN_DLC | float | Snapdragon® X Elite | 15.217 ms | 1 - 1 MB | NPU
84
+ | Beit | QNN_DLC | float | Snapdragon® 8 Gen 3 Mobile | 9.86 ms | 0 - 515 MB | NPU
85
+ | Beit | QNN_DLC | float | Qualcomm® QCS8275 (Proxy) | 47.008 ms | 1 - 358 MB | NPU
86
+ | Beit | QNN_DLC | float | Qualcomm® QCS8550 (Proxy) | 14.574 ms | 1 - 3 MB | NPU
87
+ | Beit | QNN_DLC | float | Qualcomm® SA8775P | 17.343 ms | 1 - 358 MB | NPU
88
+ | Beit | QNN_DLC | float | Qualcomm® QCS9075 | 18.669 ms | 1 - 3 MB | NPU
89
+ | Beit | QNN_DLC | float | Qualcomm® QCS8450 (Proxy) | 25.478 ms | 0 - 499 MB | NPU
90
+ | Beit | QNN_DLC | float | Qualcomm® SA7255P | 47.008 ms | 1 - 358 MB | NPU
91
+ | Beit | QNN_DLC | float | Qualcomm® SA8295P | 19.506 ms | 0 - 357 MB | NPU
92
+ | Beit | QNN_DLC | float | Snapdragon® 8 Elite For Galaxy Mobile | 8.267 ms | 0 - 359 MB | NPU
93
+ | Beit | QNN_DLC | w8a16 | Snapdragon® 8 Elite Gen 5 Mobile | 3.974 ms | 0 - 443 MB | NPU
94
+ | Beit | QNN_DLC | w8a16 | Snapdragon® X2 Elite | 4.604 ms | 0 - 0 MB | NPU
95
+ | Beit | QNN_DLC | w8a16 | Snapdragon® X Elite | 10.32 ms | 0 - 0 MB | NPU
96
+ | Beit | QNN_DLC | w8a16 | Snapdragon® 8 Gen 3 Mobile | 6.413 ms | 0 - 484 MB | NPU
97
+ | Beit | QNN_DLC | w8a16 | Qualcomm® QCS8275 (Proxy) | 19.597 ms | 0 - 415 MB | NPU
98
+ | Beit | QNN_DLC | w8a16 | Qualcomm® QCS8550 (Proxy) | 9.611 ms | 0 - 3 MB | NPU
99
+ | Beit | QNN_DLC | w8a16 | Qualcomm® SA8775P | 9.934 ms | 0 - 415 MB | NPU
100
+ | Beit | QNN_DLC | w8a16 | Qualcomm® QCS9075 | 11.607 ms | 2 - 4 MB | NPU
101
+ | Beit | QNN_DLC | w8a16 | Qualcomm® QCM6690 | 133.162 ms | 0 - 485 MB | NPU
102
+ | Beit | QNN_DLC | w8a16 | Qualcomm® SA7255P | 19.597 ms | 0 - 415 MB | NPU
103
+ | Beit | QNN_DLC | w8a16 | Snapdragon® 8 Elite For Galaxy Mobile | 5.682 ms | 0 - 391 MB | NPU
104
+ | Beit | QNN_DLC | w8a16 | Snapdragon® 7 Gen 4 Mobile | 11.418 ms | 0 - 444 MB | NPU
105
+ | Beit | TFLITE | float | Snapdragon® 8 Elite Gen 5 Mobile | 3.764 ms | 0 - 310 MB | NPU
106
+ | Beit | TFLITE | float | Snapdragon® 8 Gen 3 Mobile | 6.518 ms | 0 - 418 MB | NPU
107
+ | Beit | TFLITE | float | Qualcomm® QCS8275 (Proxy) | 37.72 ms | 0 - 314 MB | NPU
108
+ | Beit | TFLITE | float | Qualcomm® QCS8550 (Proxy) | 9.121 ms | 0 - 3 MB | NPU
109
+ | Beit | TFLITE | float | Qualcomm® SA8775P | 12.024 ms | 0 - 314 MB | NPU
110
+ | Beit | TFLITE | float | Qualcomm® QCS9075 | 12.287 ms | 0 - 187 MB | NPU
111
+ | Beit | TFLITE | float | Qualcomm® QCS8450 (Proxy) | 18.67 ms | 0 - 378 MB | NPU
112
+ | Beit | TFLITE | float | Qualcomm® SA7255P | 37.72 ms | 0 - 314 MB | NPU
113
+ | Beit | TFLITE | float | Qualcomm® SA8295P | 14.923 ms | 0 - 292 MB | NPU
114
+ | Beit | TFLITE | float | Snapdragon® 8 Elite For Galaxy Mobile | 4.519 ms | 0 - 320 MB | NPU
115
 
116
  ## License
117
  * The license for the original implementation of Beit can be found
release_assets.json CHANGED
@@ -1 +1,46 @@
1
- {"version":"0.50.2","precisions":{"float":{"universal_assets":{"onnx":{"tool_versions":{"qairt":"2.42.0.251225135753_193295","onnx_runtime":"1.24.3"},"download_url":"https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/beit/releases/v0.50.2/beit-onnx-float.zip"},"tflite":{"tool_versions":{"qairt":"2.43.0.260127150333_193827","tflite":"2.19.1"},"download_url":"https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/beit/releases/v0.50.2/beit-tflite-float.zip"}}},"w8a16":{"universal_assets":{"onnx":{"tool_versions":{"qairt":"2.42.0.251225135753_193295","onnx_runtime":"1.24.3"},"download_url":"https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/beit/releases/v0.50.2/beit-onnx-w8a16.zip"}}}}}
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "version": "0.51.0",
3
+ "precisions": {
4
+ "w8a16": {
5
+ "universal_assets": {
6
+ "qnn_dlc": {
7
+ "tool_versions": {
8
+ "qairt": "2.45.0.260326154327"
9
+ },
10
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/beit/releases/v0.51.0/beit-qnn_dlc-w8a16.zip"
11
+ },
12
+ "onnx": {
13
+ "tool_versions": {
14
+ "qairt": "2.42.0.251225135753_193295",
15
+ "onnx_runtime": "1.24.3"
16
+ },
17
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/beit/releases/v0.51.0/beit-onnx-w8a16.zip"
18
+ }
19
+ }
20
+ },
21
+ "float": {
22
+ "universal_assets": {
23
+ "tflite": {
24
+ "tool_versions": {
25
+ "qairt": "2.45.0.260326154327",
26
+ "litert": "1.4.2"
27
+ },
28
+ "download_url": "https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/beit/releases/v0.51.0/beit-tflite-float.zip"
29
+ },
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/beit/releases/v0.51.0/beit-qnn_dlc-float.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/beit/releases/v0.51.0/beit-onnx-float.zip"
42
+ }
43
+ }
44
+ }
45
+ }
46
+ }