qaihm-bot commited on
Commit
db422ab
·
verified ·
1 Parent(s): 7432cd0

See https://github.com/quic/ai-hub-models/releases/v0.42.0 for changelog.

MobileNet-v2_float.dlc CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5240d1b2fb7c913d44c81f2406c7999698e438863613debd732bdf5bf2493e5d
3
- size 14081548
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9312c43049b0933e94e62124a544f752e3dc8fb6f69d8d06e84a7718640d4a99
3
+ size 14081652
MobileNet-v2_float.onnx.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:54400575655cc5d46049526e4cf79c0fd9ad744a4479549d165cce6b1315ed62
3
- size 12978714
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc2c1ff72857727b3b603852c7d443076787ca2210d2c6aa2e7a65084b7ea792
3
+ size 12978711
MobileNet-v2_w8a16.dlc CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c9dd87593cace0086a115fc64e01c988fc1bf2c459286aad245ef8f44287eb69
3
- size 4633452
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f88602b924fa15d597c4953aaea2a5a761e8028dca424afc313e31e5662ac7d4
3
+ size 4633556
MobileNet-v2_w8a16_mixed_int16.dlc CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e66120d6b785ed88eef1203f4b5cabc1b0af54ec8216ab3c9ebf51132aa676fd
3
- size 5082812
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bed2726d0511d29e4b68059bc20d3a453c50b7ff02db67fed674ff07021fb3e3
3
+ size 5082916
README.md CHANGED
@@ -38,60 +38,63 @@ More details on model performance across various devices, can be found
38
 
39
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
40
  |---|---|---|---|---|---|---|---|---|
41
- | MobileNet-v2 | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 2.733 ms | 0 - 24 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
42
- | MobileNet-v2 | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 2.584 ms | 1 - 21 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
43
- | MobileNet-v2 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 1.04 ms | 0 - 34 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
44
- | MobileNet-v2 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 1.53 ms | 0 - 35 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
45
- | MobileNet-v2 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 0.87 ms | 0 - 69 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
46
- | MobileNet-v2 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.839 ms | 0 - 45 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
47
- | MobileNet-v2 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 0.832 ms | 0 - 46 MB | NPU | [MobileNet-v2.onnx.zip](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.onnx.zip) |
48
- | MobileNet-v2 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 1.221 ms | 0 - 25 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
49
- | MobileNet-v2 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 1.187 ms | 1 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
50
- | MobileNet-v2 | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 2.733 ms | 0 - 24 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
51
- | MobileNet-v2 | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 2.584 ms | 1 - 21 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
52
- | MobileNet-v2 | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 0.865 ms | 0 - 68 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
53
- | MobileNet-v2 | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.842 ms | 1 - 48 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
54
- | MobileNet-v2 | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 1.442 ms | 0 - 29 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
55
- | MobileNet-v2 | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 1.405 ms | 1 - 28 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
56
- | MobileNet-v2 | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 0.869 ms | 0 - 68 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
57
- | MobileNet-v2 | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.841 ms | 1 - 7 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
58
- | MobileNet-v2 | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 1.221 ms | 0 - 25 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
59
- | MobileNet-v2 | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 1.187 ms | 1 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
60
- | MobileNet-v2 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.579 ms | 0 - 43 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
61
- | MobileNet-v2 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.562 ms | 0 - 35 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
62
- | MobileNet-v2 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 0.522 ms | 0 - 36 MB | NPU | [MobileNet-v2.onnx.zip](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.onnx.zip) |
63
- | MobileNet-v2 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 0.461 ms | 0 - 28 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
64
  | MobileNet-v2 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.431 ms | 1 - 29 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
65
- | MobileNet-v2 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 0.439 ms | 0 - 24 MB | NPU | [MobileNet-v2.onnx.zip](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.onnx.zip) |
66
- | MobileNet-v2 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 0.372 ms | 0 - 27 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
67
- | MobileNet-v2 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.334 ms | 0 - 24 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
68
- | MobileNet-v2 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 0.399 ms | 1 - 28 MB | NPU | [MobileNet-v2.onnx.zip](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.onnx.zip) |
69
- | MobileNet-v2 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 1.003 ms | 53 - 53 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
70
- | MobileNet-v2 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 0.765 ms | 7 - 7 MB | NPU | [MobileNet-v2.onnx.zip](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.onnx.zip) |
71
- | MobileNet-v2 | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 1.763 ms | 0 - 21 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
72
- | MobileNet-v2 | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 0.998 ms | 0 - 33 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
73
- | MobileNet-v2 | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.81 ms | 0 - 29 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
74
- | MobileNet-v2 | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 1.037 ms | 0 - 21 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
75
- | MobileNet-v2 | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 1.763 ms | 0 - 21 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
76
- | MobileNet-v2 | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.808 ms | 0 - 28 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
77
- | MobileNet-v2 | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 1.269 ms | 0 - 27 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
78
- | MobileNet-v2 | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.809 ms | 0 - 29 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
79
- | MobileNet-v2 | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 1.037 ms | 0 - 21 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
80
- | MobileNet-v2 | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.557 ms | 0 - 32 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
81
- | MobileNet-v2 | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.368 ms | 0 - 25 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
82
- | MobileNet-v2 | w8a16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.295 ms | 0 - 27 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
83
- | MobileNet-v2 | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.947 ms | 22 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
84
- | MobileNet-v2 | w8a16_mixed_int16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 1.93 ms | 0 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
85
- | MobileNet-v2 | w8a16_mixed_int16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.853 ms | 0 - 27 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
86
- | MobileNet-v2 | w8a16_mixed_int16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 4.521 ms | 0 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
87
- | MobileNet-v2 | w8a16_mixed_int16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 1.93 ms | 0 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
88
- | MobileNet-v2 | w8a16_mixed_int16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.854 ms | 0 - 27 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
89
- | MobileNet-v2 | w8a16_mixed_int16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.855 ms | 0 - 26 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
90
- | MobileNet-v2 | w8a16_mixed_int16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 4.521 ms | 0 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
91
- | MobileNet-v2 | w8a16_mixed_int16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.594 ms | 0 - 31 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
92
- | MobileNet-v2 | w8a16_mixed_int16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.395 ms | 0 - 26 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
93
- | MobileNet-v2 | w8a16_mixed_int16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.32 ms | 0 - 28 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
94
- | MobileNet-v2 | w8a16_mixed_int16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.997 ms | 22 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
 
 
 
95
 
96
 
97
 
@@ -105,9 +108,9 @@ pip install qai-hub-models
105
  ```
106
 
107
 
108
- ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
109
 
110
- Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
111
  Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
112
 
113
  With this API token, you can configure your client to run models on the cloud
@@ -115,7 +118,7 @@ hosted devices.
115
  ```bash
116
  qai-hub configure --api_token API_TOKEN
117
  ```
118
- Navigate to [docs](https://app.aihub.qualcomm.com/docs/) for more information.
119
 
120
 
121
 
@@ -226,7 +229,7 @@ With the output of the model, you can compute like PSNR, relative errors or
226
  spot check the output with expected output.
227
 
228
  **Note**: This on-device profiling and inference requires access to Qualcomm®
229
- AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
230
 
231
 
232
 
 
38
 
39
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
40
  |---|---|---|---|---|---|---|---|---|
41
+ | MobileNet-v2 | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 2.733 ms | 0 - 25 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
42
+ | MobileNet-v2 | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 2.613 ms | 1 - 23 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
43
+ | MobileNet-v2 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 1.068 ms | 0 - 33 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
44
+ | MobileNet-v2 | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 1.538 ms | 0 - 35 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
45
+ | MobileNet-v2 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 0.881 ms | 0 - 68 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
46
+ | MobileNet-v2 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.842 ms | 1 - 47 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
47
+ | MobileNet-v2 | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 0.843 ms | 0 - 47 MB | NPU | [MobileNet-v2.onnx.zip](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.onnx.zip) |
48
+ | MobileNet-v2 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 5.121 ms | 0 - 25 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
49
+ | MobileNet-v2 | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 1.184 ms | 1 - 23 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
50
+ | MobileNet-v2 | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 2.733 ms | 0 - 25 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
51
+ | MobileNet-v2 | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 2.613 ms | 1 - 23 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
52
+ | MobileNet-v2 | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 0.872 ms | 0 - 68 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
53
+ | MobileNet-v2 | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.845 ms | 1 - 8 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
54
+ | MobileNet-v2 | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 1.433 ms | 0 - 31 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
55
+ | MobileNet-v2 | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 1.421 ms | 1 - 29 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
56
+ | MobileNet-v2 | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 0.881 ms | 0 - 68 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
57
+ | MobileNet-v2 | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.839 ms | 1 - 47 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
58
+ | MobileNet-v2 | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 5.121 ms | 0 - 25 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
59
+ | MobileNet-v2 | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 1.184 ms | 1 - 23 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
60
+ | MobileNet-v2 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 0.579 ms | 0 - 40 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
61
+ | MobileNet-v2 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.563 ms | 0 - 36 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
62
+ | MobileNet-v2 | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 0.528 ms | 0 - 38 MB | NPU | [MobileNet-v2.onnx.zip](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.onnx.zip) |
63
+ | MobileNet-v2 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 0.457 ms | 0 - 33 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
64
  | MobileNet-v2 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.431 ms | 1 - 29 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
65
+ | MobileNet-v2 | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 0.439 ms | 0 - 27 MB | NPU | [MobileNet-v2.onnx.zip](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.onnx.zip) |
66
+ | MobileNet-v2 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 0.379 ms | 0 - 28 MB | NPU | [MobileNet-v2.tflite](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.tflite) |
67
+ | MobileNet-v2 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.334 ms | 0 - 27 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
68
+ | MobileNet-v2 | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 0.393 ms | 0 - 27 MB | NPU | [MobileNet-v2.onnx.zip](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.onnx.zip) |
69
+ | MobileNet-v2 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.991 ms | 53 - 53 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.dlc) |
70
+ | MobileNet-v2 | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 0.769 ms | 7 - 7 MB | NPU | [MobileNet-v2.onnx.zip](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2.onnx.zip) |
71
+ | MobileNet-v2 | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 1.748 ms | 0 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
72
+ | MobileNet-v2 | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 0.988 ms | 0 - 34 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
73
+ | MobileNet-v2 | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.81 ms | 0 - 28 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
74
+ | MobileNet-v2 | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 1.023 ms | 0 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
75
+ | MobileNet-v2 | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN_DLC | 2.728 ms | 0 - 26 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
76
+ | MobileNet-v2 | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 1.748 ms | 0 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
77
+ | MobileNet-v2 | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.809 ms | 0 - 30 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
78
+ | MobileNet-v2 | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 1.265 ms | 0 - 28 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
79
+ | MobileNet-v2 | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.809 ms | 0 - 28 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
80
+ | MobileNet-v2 | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 1.023 ms | 0 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
81
+ | MobileNet-v2 | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.557 ms | 0 - 34 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
82
+ | MobileNet-v2 | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.371 ms | 0 - 28 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
83
+ | MobileNet-v2 | w8a16 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | QNN_DLC | 0.839 ms | 1 - 26 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
84
+ | MobileNet-v2 | w8a16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.295 ms | 0 - 29 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
85
+ | MobileNet-v2 | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 0.949 ms | 21 - 21 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16.dlc) |
86
+ | MobileNet-v2 | w8a16_mixed_int16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 1.911 ms | 0 - 23 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
87
+ | MobileNet-v2 | w8a16_mixed_int16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 0.858 ms | 0 - 26 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
88
+ | MobileNet-v2 | w8a16_mixed_int16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 4.505 ms | 0 - 23 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
89
+ | MobileNet-v2 | w8a16_mixed_int16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 1.911 ms | 0 - 23 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
90
+ | MobileNet-v2 | w8a16_mixed_int16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 0.861 ms | 0 - 26 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
91
+ | MobileNet-v2 | w8a16_mixed_int16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 0.86 ms | 0 - 27 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
92
+ | MobileNet-v2 | w8a16_mixed_int16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 4.505 ms | 0 - 23 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
93
+ | MobileNet-v2 | w8a16_mixed_int16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 0.601 ms | 0 - 37 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
94
+ | MobileNet-v2 | w8a16_mixed_int16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 0.394 ms | 0 - 27 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
95
+ | MobileNet-v2 | w8a16_mixed_int16 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | QNN_DLC | 0.942 ms | 0 - 30 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
96
+ | MobileNet-v2 | w8a16_mixed_int16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 0.32 ms | 0 - 30 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
97
+ | MobileNet-v2 | w8a16_mixed_int16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 1.015 ms | 22 - 22 MB | NPU | [MobileNet-v2.dlc](https://huggingface.co/qualcomm/MobileNet-v2/blob/main/MobileNet-v2_w8a16_mixed_int16.dlc) |
98
 
99
 
100
 
 
108
  ```
109
 
110
 
111
+ ## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
112
 
113
+ Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.com/) with your
114
  Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
115
 
116
  With this API token, you can configure your client to run models on the cloud
 
118
  ```bash
119
  qai-hub configure --api_token API_TOKEN
120
  ```
121
+ Navigate to [docs](https://workbench.aihub.qualcomm.com/docs/) for more information.
122
 
123
 
124
 
 
229
  spot check the output with expected output.
230
 
231
  **Note**: This on-device profiling and inference requires access to Qualcomm®
232
+ AI Hub Workbench. [Sign up for access](https://myaccount.qualcomm.com/signup).
233
 
234
 
235
 
tool-versions.yaml CHANGED
@@ -1,3 +1,3 @@
1
  tool_versions:
2
  qnn_dlc:
3
- qairt: 2.39.0.250925215840_163802
 
1
  tool_versions:
2
  qnn_dlc:
3
+ qairt: 2.40.0.251030114326_189385