qaihm-bot commited on
Commit
dcab83b
·
verified ·
1 Parent(s): 8e37a43

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

README.md CHANGED
@@ -39,89 +39,94 @@ More details on model performance across various devices, can be found
39
 
40
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
41
  |---|---|---|---|---|---|---|---|---|
42
- | Segformer-Base | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 216.712 ms | 6 - 55 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
43
- | Segformer-Base | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 210.449 ms | 2 - 43 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
44
- | Segformer-Base | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 124.211 ms | 10 - 76 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
45
- | Segformer-Base | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 117.429 ms | 3 - 54 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
46
- | Segformer-Base | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 110.866 ms | 10 - 27 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
47
- | Segformer-Base | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 108.127 ms | 3 - 18 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
48
- | Segformer-Base | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 109.074 ms | 10 - 41 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
49
- | Segformer-Base | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 113.544 ms | 9 - 59 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
50
- | Segformer-Base | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 110.236 ms | 2 - 44 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
51
- | Segformer-Base | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 216.712 ms | 6 - 55 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
52
- | Segformer-Base | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 210.449 ms | 2 - 43 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
53
- | Segformer-Base | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 111.233 ms | 9 - 28 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
54
- | Segformer-Base | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 108.776 ms | 3 - 18 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
55
- | Segformer-Base | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 136.454 ms | 9 - 75 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
56
- | Segformer-Base | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 119.877 ms | 0 - 49 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
57
- | Segformer-Base | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 111.087 ms | 9 - 27 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
58
- | Segformer-Base | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 108.076 ms | 3 - 18 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
59
- | Segformer-Base | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 113.544 ms | 9 - 59 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
60
- | Segformer-Base | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 110.236 ms | 2 - 44 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
61
- | Segformer-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 83.415 ms | 8 - 66 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
62
- | Segformer-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 81.256 ms | 3 - 57 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
63
- | Segformer-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 84.243 ms | 24 - 79 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
64
- | Segformer-Base | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 19.81 ms | 9 - 195 MB | GPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
65
- | Segformer-Base | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 73.275 ms | 3 - 51 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
66
- | Segformer-Base | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 74.664 ms | 44 - 92 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
67
- | Segformer-Base | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 74.641 ms | 8 - 61 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
68
- | Segformer-Base | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 74.27 ms | 3 - 57 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
69
- | Segformer-Base | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 74.095 ms | 23 - 75 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
70
- | Segformer-Base | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 113.245 ms | 3 - 3 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
71
- | Segformer-Base | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 112.96 ms | 33 - 33 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
72
- | Segformer-Base | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 38.707 ms | 2 - 63 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
73
- | Segformer-Base | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 26.004 ms | 0 - 25 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
74
- | Segformer-Base | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 62.81 ms | 62 - 74 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
75
- | Segformer-Base | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 25.186 ms | 2 - 60 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
76
- | Segformer-Base | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | ONNX | 456.698 ms | 374 - 390 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
77
- | Segformer-Base | w8a16 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | ONNX | 378.061 ms | 375 - 380 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
78
- | Segformer-Base | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 38.707 ms | 2 - 63 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
79
- | Segformer-Base | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 25.539 ms | 0 - 24 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
80
- | Segformer-Base | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 25.682 ms | 0 - 27 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
81
- | Segformer-Base | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 25.186 ms | 2 - 60 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
82
- | Segformer-Base | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 19.294 ms | 2 - 68 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
83
- | Segformer-Base | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 51.756 ms | 63 - 99 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
84
- | Segformer-Base | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 15.185 ms | 2 - 63 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
85
- | Segformer-Base | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 45.799 ms | 66 - 94 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
86
- | Segformer-Base | w8a16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 13.233 ms | 2 - 56 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
87
- | Segformer-Base | w8a16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 40.015 ms | 62 - 94 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
88
- | Segformer-Base | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 24.022 ms | 39 - 39 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
89
- | Segformer-Base | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 55.357 ms | 131 - 131 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
90
- | Segformer-Base | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 21.301 ms | 2 - 39 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
91
- | Segformer-Base | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 12.21 ms | 1 - 37 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
92
- | Segformer-Base | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 12.414 ms | 2 - 50 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
93
- | Segformer-Base | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 7.889 ms | 1 - 52 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
94
- | Segformer-Base | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 11.763 ms | 2 - 13 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
95
- | Segformer-Base | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 6.604 ms | 1 - 12 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
96
- | Segformer-Base | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 55.622 ms | 52 - 65 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
97
- | Segformer-Base | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 12.419 ms | 2 - 39 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
98
- | Segformer-Base | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 7.236 ms | 1 - 37 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
99
- | Segformer-Base | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | TFLITE | 122.189 ms | 15 - 51 MB | GPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
100
- | Segformer-Base | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN_DLC | 25.923 ms | 1 - 60 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
101
- | Segformer-Base | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | ONNX | 213.013 ms | 196 - 211 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
102
- | Segformer-Base | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 381.814 ms | 0 - 38 MB | CPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
103
- | Segformer-Base | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | ONNX | 188.732 ms | 195 - 202 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
104
- | Segformer-Base | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 21.301 ms | 2 - 39 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
105
- | Segformer-Base | w8a8 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 12.21 ms | 1 - 37 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
106
- | Segformer-Base | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 11.788 ms | 2 - 18 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
107
- | Segformer-Base | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 6.605 ms | 1 - 12 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
108
- | Segformer-Base | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 14.464 ms | 2 - 47 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
109
- | Segformer-Base | w8a8 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 8.598 ms | 1 - 45 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
110
- | Segformer-Base | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 11.794 ms | 2 - 14 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
111
- | Segformer-Base | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 6.604 ms | 1 - 11 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
112
- | Segformer-Base | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 12.419 ms | 2 - 39 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
113
- | Segformer-Base | w8a8 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 7.236 ms | 1 - 37 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
114
- | Segformer-Base | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 8.132 ms | 2 - 50 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
115
- | Segformer-Base | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 4.431 ms | 1 - 53 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
116
- | Segformer-Base | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 44.222 ms | 57 - 89 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
117
- | Segformer-Base | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 6.842 ms | 1 - 42 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
118
- | Segformer-Base | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 3.307 ms | 1 - 44 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
119
- | Segformer-Base | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 40.723 ms | 59 - 87 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
120
- | Segformer-Base | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 6.26 ms | 2 - 45 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
121
- | Segformer-Base | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 2.753 ms | 1 - 46 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
122
- | Segformer-Base | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 36.221 ms | 53 - 82 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
123
- | Segformer-Base | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 7.183 ms | 14 - 14 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
124
- | Segformer-Base | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 45.314 ms | 65 - 65 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
 
 
 
 
 
125
 
126
 
127
 
@@ -135,9 +140,9 @@ pip install qai-hub-models
135
  ```
136
 
137
 
138
- ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
139
 
140
- Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
141
  Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
142
 
143
  With this API token, you can configure your client to run models on the cloud
@@ -145,7 +150,7 @@ hosted devices.
145
  ```bash
146
  qai-hub configure --api_token API_TOKEN
147
  ```
148
- Navigate to [docs](https://app.aihub.qualcomm.com/docs/) for more information.
149
 
150
 
151
 
@@ -256,7 +261,7 @@ With the output of the model, you can compute like PSNR, relative errors or
256
  spot check the output with expected output.
257
 
258
  **Note**: This on-device profiling and inference requires access to Qualcomm®
259
- AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).
260
 
261
 
262
 
 
39
 
40
  | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
41
  |---|---|---|---|---|---|---|---|---|
42
+ | Segformer-Base | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 216.689 ms | 9 - 58 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
43
+ | Segformer-Base | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 212.355 ms | 3 - 43 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
44
+ | Segformer-Base | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 112.68 ms | 9 - 67 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
45
+ | Segformer-Base | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 124.566 ms | 3 - 50 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
46
+ | Segformer-Base | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 111.227 ms | 9 - 27 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
47
+ | Segformer-Base | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 109.185 ms | 3 - 20 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
48
+ | Segformer-Base | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 108.782 ms | 19 - 51 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
49
+ | Segformer-Base | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 532.616 ms | 8 - 56 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
50
+ | Segformer-Base | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 111.577 ms | 1 - 41 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
51
+ | Segformer-Base | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 216.689 ms | 9 - 58 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
52
+ | Segformer-Base | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 212.355 ms | 3 - 43 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
53
+ | Segformer-Base | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 111.612 ms | 10 - 27 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
54
+ | Segformer-Base | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 109.235 ms | 3 - 18 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
55
+ | Segformer-Base | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 122.788 ms | 9 - 64 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
56
+ | Segformer-Base | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 124.886 ms | 3 - 49 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
57
+ | Segformer-Base | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 111.388 ms | 9 - 27 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
58
+ | Segformer-Base | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 109.107 ms | 3 - 19 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
59
+ | Segformer-Base | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 532.616 ms | 8 - 56 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
60
+ | Segformer-Base | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 111.577 ms | 1 - 41 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
61
+ | Segformer-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 83.609 ms | 9 - 69 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
62
+ | Segformer-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 83.663 ms | 3 - 54 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
63
+ | Segformer-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 84.336 ms | 25 - 84 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
64
+ | Segformer-Base | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 74.221 ms | 9 - 64 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
65
+ | Segformer-Base | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 74.772 ms | 3 - 49 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
66
+ | Segformer-Base | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 74.762 ms | 21 - 69 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
67
+ | Segformer-Base | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 73.692 ms | 8 - 60 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
68
+ | Segformer-Base | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 73.766 ms | 2 - 51 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
69
+ | Segformer-Base | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 74.15 ms | 23 - 73 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
70
+ | Segformer-Base | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 113.856 ms | 23 - 23 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
71
+ | Segformer-Base | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 112.849 ms | 33 - 33 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
72
+ | Segformer-Base | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 37.205 ms | 2 - 49 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
73
+ | Segformer-Base | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 24.646 ms | 2 - 18 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
74
+ | Segformer-Base | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 62.796 ms | 62 - 75 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
75
+ | Segformer-Base | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 24.217 ms | 2 - 49 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
76
+ | Segformer-Base | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | ONNX | 453.946 ms | 374 - 390 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
77
+ | Segformer-Base | w8a16 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | ONNX | 426.941 ms | 369 - 374 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
78
+ | Segformer-Base | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 37.205 ms | 2 - 49 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
79
+ | Segformer-Base | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 24.523 ms | 0 - 16 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
80
+ | Segformer-Base | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 25.296 ms | 2 - 18 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
81
+ | Segformer-Base | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 24.217 ms | 2 - 49 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
82
+ | Segformer-Base | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 18.687 ms | 2 - 57 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
83
+ | Segformer-Base | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 49.515 ms | 62 - 96 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
84
+ | Segformer-Base | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 14.156 ms | 2 - 53 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
85
+ | Segformer-Base | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 45.554 ms | 65 - 94 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
86
+ | Segformer-Base | w8a16 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | QNN_DLC | 45.867 ms | 2 - 69 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
87
+ | Segformer-Base | w8a16 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | ONNX | 412.713 ms | 364 - 382 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
88
+ | Segformer-Base | w8a16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 13.146 ms | 2 - 59 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
89
+ | Segformer-Base | w8a16 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 40.063 ms | 62 - 94 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
90
+ | Segformer-Base | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 23.319 ms | 15 - 15 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
91
+ | Segformer-Base | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 52.567 ms | 131 - 131 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
92
+ | Segformer-Base | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 21.178 ms | 2 - 41 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
93
+ | Segformer-Base | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 10.974 ms | 1 - 39 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
94
+ | Segformer-Base | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 12.553 ms | 2 - 50 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
95
+ | Segformer-Base | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 8.656 ms | 1 - 51 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
96
+ | Segformer-Base | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 11.605 ms | 0 - 12 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
97
+ | Segformer-Base | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 5.871 ms | 1 - 12 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
98
+ | Segformer-Base | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 57.051 ms | 52 - 70 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
99
+ | Segformer-Base | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 54.878 ms | 2 - 41 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
100
+ | Segformer-Base | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 6.548 ms | 1 - 40 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
101
+ | Segformer-Base | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | TFLITE | 137.324 ms | 15 - 55 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
102
+ | Segformer-Base | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN_DLC | 21.665 ms | 0 - 50 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
103
+ | Segformer-Base | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | ONNX | 211.012 ms | 196 - 212 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
104
+ | Segformer-Base | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 372.021 ms | 5 - 43 MB | CPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
105
+ | Segformer-Base | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | ONNX | 186.882 ms | 197 - 203 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
106
+ | Segformer-Base | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 21.178 ms | 2 - 41 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
107
+ | Segformer-Base | w8a8 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 10.974 ms | 1 - 39 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
108
+ | Segformer-Base | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 11.66 ms | 2 - 17 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
109
+ | Segformer-Base | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 5.89 ms | 1 - 12 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
110
+ | Segformer-Base | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 14.617 ms | 2 - 47 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
111
+ | Segformer-Base | w8a8 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 7.915 ms | 1 - 48 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
112
+ | Segformer-Base | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 11.68 ms | 0 - 10 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
113
+ | Segformer-Base | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 5.87 ms | 1 - 11 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
114
+ | Segformer-Base | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 54.878 ms | 2 - 41 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
115
+ | Segformer-Base | w8a8 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 6.548 ms | 1 - 40 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
116
+ | Segformer-Base | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 8.073 ms | 2 - 52 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
117
+ | Segformer-Base | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 4.013 ms | 1 - 49 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
118
+ | Segformer-Base | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 47.036 ms | 57 - 88 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
119
+ | Segformer-Base | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 6.474 ms | 2 - 44 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
120
+ | Segformer-Base | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 2.882 ms | 1 - 43 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
121
+ | Segformer-Base | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 41.222 ms | 59 - 87 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
122
+ | Segformer-Base | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | TFLITE | 47.239 ms | 15 - 44 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
123
+ | Segformer-Base | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | QNN_DLC | 6.229 ms | 1 - 44 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
124
+ | Segformer-Base | w8a8 | Snapdragon 7 Gen 4 QRD | Snapdragon® 7 Gen 4 Mobile | ONNX | 177.434 ms | 183 - 201 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
125
+ | Segformer-Base | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 6.092 ms | 2 - 47 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
126
+ | Segformer-Base | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 2.315 ms | 1 - 45 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
127
+ | Segformer-Base | w8a8 | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 35.932 ms | 53 - 83 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
128
+ | Segformer-Base | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 6.425 ms | 18 - 18 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.dlc) |
129
+ | Segformer-Base | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 50.119 ms | 65 - 65 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
130
 
131
 
132
 
 
140
  ```
141
 
142
 
143
+ ## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
144
 
145
+ Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.com/) with your
146
  Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.
147
 
148
  With this API token, you can configure your client to run models on the cloud
 
150
  ```bash
151
  qai-hub configure --api_token API_TOKEN
152
  ```
153
+ Navigate to [docs](https://workbench.aihub.qualcomm.com/docs/) for more information.
154
 
155
 
156
 
 
261
  spot check the output with expected output.
262
 
263
  **Note**: This on-device profiling and inference requires access to Qualcomm®
264
+ AI Hub Workbench. [Sign up for access](https://myaccount.qualcomm.com/signup).
265
 
266
 
267
 
Segformer-Base_float.dlc CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e58d6e0d5d2b8af03b2bd626732985325a7f323e67473de69514e6f034d1ab4b
3
- size 15339348
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70878dbeb930936ac2a84d84ce90c1f109a37e15cfb225865217e9724c80bfc4
3
+ size 15339444
Segformer-Base_float.onnx.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2a1261a396f236438909c9fd64650057f9d5d48b28921e93327b56dadef04cd7
3
- size 13994868
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27f87e3f0eb94fd987bf454d0395df5770df572627d532ecde9dc4f81c9b04b1
3
+ size 13994858
Segformer-Base_w8a16.dlc CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:85383671e8fcb88a52e39311ee2a768c6808f661dd16d625f484c67fdfbfa90f
3
- size 4753492
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38348d54819b6c524004de5cb2d345870f1048dad17511b129fb275531abc452
3
+ size 4753596
Segformer-Base_w8a16.onnx.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9debd1ad86b55d63590456205b4eaf11ad11d46d1613ab3fd2885271a98ac037
3
- size 10813980
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c468a41eb54eb851056e31f6caf580ca4b57f72dcc6732557a0368199fd900f8
3
+ size 10813967
Segformer-Base_w8a8.dlc CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c0ac15fee535654799729ca741a8665cac5937ec4c4a64bafcb7d7429cfe8d33
3
- size 4751948
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:235963a3b8645467b40142a86db72f058fe5f847fd3936b3e46b2e9d1aca8ab9
3
+ size 4752052
Segformer-Base_w8a8.onnx.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:69d8435566169ad9fa4a4c4332c5e4f652e79380dd6d38795d478e490f6b020c
3
- size 10817596
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7862a9bd71f3a8aac49c345b57c14b92a55642bfb9be5c00c3118fd614f8c2b7
3
+ size 10817609
Segformer-Base_w8a8.tflite CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:46db52a2841444e85700eb54e91412c292c661b8337849b9994bfa35d36fa0b1
3
  size 4087672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:747c2c4b8c38e21ffaba175415d12000eb3ef581306053d70b6138823959d020
3
  size 4087672