qaihm-bot commited on
Commit
c7592a7
·
verified ·
1 Parent(s): 0c97370

See https://github.com/quic/ai-hub-models/releases/v0.38.0 for changelog.

README.md CHANGED
@@ -39,64 +39,68 @@ 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 | 229.458 ms | 10 - 62 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 | 210.644 ms | 2 - 42 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 | 117.21 ms | 9 - 68 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 | 116.416 ms | 3 - 53 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 | 118.763 ms | 0 - 30 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 | 108.077 ms | 3 - 15 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 | 109.34 ms | 19 - 50 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 | 121.107 ms | 9 - 62 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 | 110.323 ms | 0 - 40 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 | 229.458 ms | 10 - 62 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 | 210.644 ms | 2 - 42 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 | 118.532 ms | 0 - 30 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 | 108.373 ms | 4 - 17 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 | 128.359 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 | 120.009 ms | 0 - 48 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 | 118.536 ms | 0 - 28 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 | 108.124 ms | 4 - 20 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 | 121.107 ms | 9 - 62 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 | 110.323 ms | 0 - 40 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
61
- | Segformer-Base | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 118.934 ms | 0 - 22 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
62
- | Segformer-Base | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_DLC | 108.128 ms | 3 - 18 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
63
- | Segformer-Base | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 109.194 ms | 19 - 51 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
64
- | Segformer-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 89.104 ms | 9 - 71 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
65
- | Segformer-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 81.639 ms | 3 - 53 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
66
- | Segformer-Base | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 82.15 ms | 26 - 77 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 QRD | Snapdragon® 8 Elite Mobile | TFLITE | 91.782 ms | 9 - 62 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.tflite) |
68
- | Segformer-Base | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_DLC | 74.695 ms | 3 - 51 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
69
- | Segformer-Base | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 76.969 ms | 25 - 69 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 | 113.369 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 | 116.566 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 | 26.486 ms | 2 - 39 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
73
- | Segformer-Base | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 20.305 ms | 2 - 53 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) | QNN_DLC | 15.333 ms | 2 - 14 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
75
- | Segformer-Base | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 16.024 ms | 2 - 40 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) | QNN_DLC | 55.407 ms | 2 - 82 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
77
- | Segformer-Base | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 26.486 ms | 2 - 39 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
78
- | Segformer-Base | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 15.3 ms | 2 - 14 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
79
- | Segformer-Base | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 19.269 ms | 2 - 52 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 | 15.305 ms | 2 - 14 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 | 16.024 ms | 2 - 40 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
82
- | Segformer-Base | w8a16 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_DLC | 15.365 ms | 0 - 15 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 | QNN_DLC | 10.303 ms | 2 - 51 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
84
- | Segformer-Base | w8a16 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_DLC | 9.381 ms | 2 - 47 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
85
- | Segformer-Base | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 15.95 ms | 1 - 1 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
86
- | Segformer-Base | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 21.496 ms | 2 - 39 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
87
- | Segformer-Base | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 12.728 ms | 2 - 49 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
88
- | Segformer-Base | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 11.858 ms | 2 - 16 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
89
- | Segformer-Base | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 12.474 ms | 2 - 40 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
90
- | Segformer-Base | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | TFLITE | 123.442 ms | 15 - 52 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
91
- | Segformer-Base | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 391.41 ms | 1 - 39 MB | CPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
92
- | Segformer-Base | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 21.496 ms | 2 - 39 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
93
- | Segformer-Base | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 11.849 ms | 2 - 18 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
94
- | Segformer-Base | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 14.773 ms | 2 - 47 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
95
- | Segformer-Base | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 11.855 ms | 2 - 17 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
96
- | Segformer-Base | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 12.474 ms | 2 - 40 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
97
- | Segformer-Base | w8a8 | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 11.869 ms | 2 - 16 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
98
- | Segformer-Base | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 8.2 ms | 2 - 47 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
99
- | Segformer-Base | w8a8 | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 6.98 ms | 1 - 42 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
 
 
 
 
100
 
101
 
102
 
@@ -178,7 +182,7 @@ from qai_hub_models.models.segformer_base import Model
178
  torch_model = Model.from_pretrained()
179
 
180
  # Device
181
- device = hub.Device("Samsung Galaxy S24")
182
 
183
  # Trace model
184
  input_shape = torch_model.get_input_spec()
 
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 | 216.605 ms | 8 - 57 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 | 210.564 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 | 123.972 ms | 9 - 82 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 | 116.901 ms | 3 - 56 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 | 110.63 ms | 10 - 28 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 | 108.135 ms | 3 - 19 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 | 109.716 ms | 19 - 50 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 | 113.257 ms | 9 - 58 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 | 110.23 ms | 2 - 43 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 | 216.605 ms | 8 - 57 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 | 210.564 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 | 111.018 ms | 10 - 30 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 | 108.158 ms | 3 - 19 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 | 136.293 ms | 9 - 80 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 | 119.953 ms | 2 - 51 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 | 110.688 ms | 9 - 26 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 | 108.405 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 | 113.257 ms | 9 - 58 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 | 110.23 ms | 2 - 43 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 | 82.893 ms | 8 - 67 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 | 82.137 ms | 3 - 52 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 | 82.757 ms | 22 - 77 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 | 74.565 ms | 8 - 63 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 | 73.303 ms | 3 - 50 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 | 76.69 ms | 21 - 66 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
67
+ | Segformer-Base | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 113.206 ms | 3 - 3 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.dlc) |
68
+ | Segformer-Base | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 115.388 ms | 33 - 33 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base.onnx.zip) |
69
+ | Segformer-Base | w8a16 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 27.083 ms | 1 - 44 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
70
+ | Segformer-Base | w8a16 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 20.507 ms | 2 - 50 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
71
+ | Segformer-Base | w8a16 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 15.768 ms | 1 - 16 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
72
+ | Segformer-Base | w8a16 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 16.142 ms | 2 - 40 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
73
+ | Segformer-Base | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | QNN_DLC | 55.524 ms | 2 - 83 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
74
+ | Segformer-Base | w8a16 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | ONNX | 450.669 ms | 375 - 390 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
75
+ | Segformer-Base | w8a16 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | ONNX | 382.425 ms | 368 - 376 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
76
+ | Segformer-Base | w8a16 | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 27.083 ms | 1 - 44 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
77
+ | Segformer-Base | w8a16 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 15.738 ms | 1 - 15 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
78
+ | Segformer-Base | w8a16 | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 19.454 ms | 2 - 51 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
79
+ | Segformer-Base | w8a16 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 15.778 ms | 1 - 16 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
80
+ | Segformer-Base | w8a16 | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 16.142 ms | 2 - 40 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
81
+ | Segformer-Base | w8a16 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 10.325 ms | 2 - 54 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
82
+ | Segformer-Base | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 8.058 ms | 2 - 53 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
83
+ | Segformer-Base | w8a16 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 35.7 ms | 38 - 521 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.onnx.zip) |
84
+ | Segformer-Base | w8a16 | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 16.198 ms | 2 - 2 MB | NPU | [Segformer-Base.dlc](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a16.dlc) |
85
+ | Segformer-Base | w8a8 | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 21.325 ms | 2 - 39 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
86
+ | Segformer-Base | w8a8 | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 12.235 ms | 2 - 49 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
87
+ | Segformer-Base | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 11.759 ms | 2 - 15 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
88
+ | Segformer-Base | w8a8 | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 64.493 ms | 6 - 101 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
89
+ | Segformer-Base | w8a8 | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 12.369 ms | 2 - 40 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
90
+ | Segformer-Base | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | TFLITE | 137.92 ms | 15 - 52 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
91
+ | Segformer-Base | w8a8 | RB3 Gen 2 (Proxy) | Qualcomm® QCS6490 (Proxy) | ONNX | 273.902 ms | 227 - 243 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
92
+ | Segformer-Base | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | TFLITE | 414.071 ms | 4 - 42 MB | CPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
93
+ | Segformer-Base | w8a8 | RB5 (Proxy) | Qualcomm® QCS8250 (Proxy) | ONNX | 237.072 ms | 226 - 238 MB | CPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
94
+ | Segformer-Base | w8a8 | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 21.325 ms | 2 - 39 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
95
+ | Segformer-Base | w8a8 | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 11.777 ms | 2 - 14 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
96
+ | Segformer-Base | w8a8 | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 14.376 ms | 2 - 47 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
97
+ | Segformer-Base | w8a8 | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 11.773 ms | 2 - 15 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
98
+ | Segformer-Base | w8a8 | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 12.369 ms | 2 - 40 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
99
+ | Segformer-Base | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 8.159 ms | 1 - 48 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
100
+ | Segformer-Base | w8a8 | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 49.693 ms | 13 - 241 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
101
+ | Segformer-Base | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 6.765 ms | 2 - 45 MB | NPU | [Segformer-Base.tflite](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.tflite) |
102
+ | Segformer-Base | w8a8 | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 33.557 ms | 24 - 242 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
103
+ | Segformer-Base | w8a8 | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 65.176 ms | 29 - 29 MB | NPU | [Segformer-Base.onnx.zip](https://huggingface.co/qualcomm/Segformer-Base/blob/main/Segformer-Base_w8a8.onnx.zip) |
104
 
105
 
106
 
 
182
  torch_model = Model.from_pretrained()
183
 
184
  # Device
185
+ device = hub.Device("Samsung Galaxy S25")
186
 
187
  # Trace model
188
  input_shape = torch_model.get_input_spec()
Segformer-Base_float.dlc CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:12a0d2cef52dae5c364093d4e75bccffec4dcab2523e00e8cd259058c51ed24f
3
- size 15337724
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:617d6e1f93bc5873006fd7772b709421fc6cbfd67e0381873d68ad8fc58137aa
3
+ size 15337812
Segformer-Base_float.onnx.zip CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d7903af853993d75201776a5e232e7b08ab45a3b915b4475213d631ff6619bbb
3
- size 13994699
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cbe0a93759080818e55bc679c960677fc2827e875e1c7734d109874db9916039
3
+ size 13994868
Segformer-Base_w8a16.dlc CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:550d3970323bf37c641c7ab0f02e7b72a6756dbe771973d13a968afac9da522f
3
- size 4762932
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d374fa90cfad5298246f0226249be5fccf1a880b7943158f0ae39d088b74242
3
+ size 4762948
precompiled/qualcomm-qcs6490-proxy/Segformer-Base_w8a16.bin → Segformer-Base_w8a16.onnx.zip RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7aa7ec97bbc118e95f81003af2ccffda8876d204780f976ae2078aafd46f674a
3
- size 6725632
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cc924f039b53bb6e72dce8da661dc17e450e8541b600b161e5fae8848434fca7
3
+ size 10812758
precompiled/qualcomm-snapdragon-x-elite/Segformer-Base_w8a16.bin → Segformer-Base_w8a8.onnx.zip RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8e11a1a14c609161e90ec662e5048bcd7060755337b1cc3887dfb980965fea46
3
- size 5279744
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:854bcf4b05906b8bee0f42d7ba513e757d5205acfd5970942168986dba4c0a1e
3
+ size 10816605
Segformer-Base_w8a8.tflite CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c090ab295173c96fc1bd4c3a91c3fc23171791c49e125ec08878e87777e97edd
3
  size 4087672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56055195d3a2daa164e034ec3e3672557fcbd5c5280a9c9c747d70da5e8b53e9
3
  size 4087672
precompiled/qualcomm-qcs6490-proxy/tool-versions.yaml DELETED
@@ -1,3 +0,0 @@
1
- tool_versions:
2
- qnn_context_binary:
3
- qairt: 2.37.0.250724175447_124859
 
 
 
 
precompiled/qualcomm-snapdragon-x-elite/Segformer-Base_float.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:be1f7f72fb0a06312204bedede4970fe802933bad43f406d800311931bee0231
3
- size 8732672
 
 
 
 
precompiled/qualcomm-snapdragon-x-elite/Segformer-Base_float.onnx.zip DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:8cc0eab2f8f25058053770c537d78d885ba93bf37e21b68c103b600901fc88ab
3
- size 7233185
 
 
 
 
precompiled/qualcomm-snapdragon-x-elite/tool-versions.yaml DELETED
@@ -1,3 +0,0 @@
1
- tool_versions:
2
- qnn_context_binary:
3
- qairt: 2.37.0.250724175447_124859
 
 
 
 
tool-versions.yaml CHANGED
@@ -1,4 +1,4 @@
1
  tool_versions:
2
- tflite:
3
- qairt: 2.37.0.250724175447_124859
4
- tflite: 2.17.0
 
1
  tool_versions:
2
+ onnx:
3
+ qairt: 2.37.1.250807093845_124904
4
+ onnx_runtime: 1.22.2