Hugging Face's logo

qualcomm
/
XLSR-Quantized

Image-to-Image
PyTorch
TF Lite
ONNX
quantized
android
Model card Files Community
2
qaihm-bot commited on
Commit
e8dc4dc
·
verified ·
1 Parent(s): 5c80179

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +34 -34
README.md CHANGED
@@ -35,48 +35,47 @@ More details on model performance across various devices, can be found
35
 
36
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
- | XLSR-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 1.092 ms | 0 - 11 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
39
- | XLSR-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 0.552 ms | 0 - 11 MB | INT8 | NPU | [XLSR-Quantized.so](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.so) |
40
- | XLSR-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 1.483 ms | 0 - 9 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
41
- | XLSR-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 0.886 ms | 0 - 24 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
42
- | XLSR-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 0.382 ms | 0 - 24 MB | INT8 | NPU | [XLSR-Quantized.so](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.so) |
43
- | XLSR-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 1.045 ms | 0 - 24 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
44
- | XLSR-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 0.869 ms | 0 - 17 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
45
- | XLSR-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 0.407 ms | 0 - 17 MB | INT8 | NPU | Use Export Script |
46
- | XLSR-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 0.894 ms | 0 - 18 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
47
- | XLSR-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 2.826 ms | 0 - 14 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
48
- | XLSR-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 1.125 ms | 0 - 12 MB | INT8 | NPU | Use Export Script |
49
- | XLSR-Quantized | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 16.291 ms | 3 - 12 MB | INT8 | GPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
50
- | XLSR-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.184 ms | 0 - 11 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
51
- | XLSR-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 0.446 ms | 0 - 3 MB | INT8 | NPU | Use Export Script |
52
- | XLSR-Quantized | SA7255P ADP | SA7255P | TFLITE | 4.328 ms | 2 - 12 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
53
- | XLSR-Quantized | SA7255P ADP | SA7255P | QNN | 2.539 ms | 0 - 11 MB | INT8 | NPU | Use Export Script |
54
- | XLSR-Quantized | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.227 ms | 1 - 11 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
55
- | XLSR-Quantized | SA8255 (Proxy) | SA8255P Proxy | QNN | 0.446 ms | 0 - 3 MB | INT8 | NPU | Use Export Script |
56
- | XLSR-Quantized | SA8295P ADP | SA8295P | TFLITE | 1.959 ms | 2 - 18 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
57
- | XLSR-Quantized | SA8295P ADP | SA8295P | QNN | 1.056 ms | 0 - 14 MB | INT8 | NPU | Use Export Script |
58
- | XLSR-Quantized | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.067 ms | 0 - 10 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
59
- | XLSR-Quantized | SA8650 (Proxy) | SA8650P Proxy | QNN | 0.449 ms | 0 - 3 MB | INT8 | NPU | Use Export Script |
60
- | XLSR-Quantized | SA8775P ADP | SA8775P | TFLITE | 2.135 ms | 1 - 11 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
61
- | XLSR-Quantized | SA8775P ADP | SA8775P | QNN | 0.95 ms | 0 - 10 MB | INT8 | NPU | Use Export Script |
62
- | XLSR-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 1.86 ms | 0 - 20 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
63
- | XLSR-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 0.677 ms | 0 - 16 MB | INT8 | NPU | Use Export Script |
64
- | XLSR-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 3.146 ms | 0 - 0 MB | INT8 | NPU | Use Export Script |
65
- | XLSR-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.54 ms | 9 - 9 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
66
 
67
 
68
 
69
 
70
  ## Installation
71
 
72
- This model can be installed as a Python package via pip.
73
 
 
74
  ```bash
75
- pip install "qai-hub-models[xlsr_quantized]"
76
  ```
77
 
78
 
79
-
80
  ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
81
 
82
  Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
@@ -128,7 +127,7 @@ XLSR-Quantized
128
  Device : Samsung Galaxy S23 (13)
129
  Runtime : TFLITE
130
  Estimated inference time (ms) : 1.1
131
- Estimated peak memory usage (MB): [0, 11]
132
  Total # Ops : 19
133
  Compute Unit(s) : NPU (16 ops) CPU (3 ops)
134
  ```
@@ -155,7 +154,7 @@ from qai_hub_models.models.xlsr_quantized import Model
155
  torch_model = Model.from_pretrained()
156
 
157
  # Device
158
- device = hub.Device("Samsung Galaxy S23")
159
 
160
  # Trace model
161
  input_shape = torch_model.get_input_spec()
@@ -247,7 +246,8 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
247
 
248
 
249
  ## License
250
- * The license for the original implementation of XLSR-Quantized can be found [here](https://github.com/quic/aimet-model-zoo/blob/develop/LICENSE.pdf).
 
251
  * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
252
 
253
 
 
35
 
36
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
37
  |---|---|---|---|---|---|---|---|---|
38
+ | XLSR-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 1.092 ms | 0 - 10 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
39
+ | XLSR-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 0.558 ms | 0 - 12 MB | INT8 | NPU | [XLSR-Quantized.so](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.so) |
40
+ | XLSR-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 1.477 ms | 0 - 9 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
41
+ | XLSR-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 0.885 ms | 0 - 21 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
42
+ | XLSR-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 0.378 ms | 0 - 22 MB | INT8 | NPU | [XLSR-Quantized.so](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.so) |
43
+ | XLSR-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 1.033 ms | 0 - 22 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
44
+ | XLSR-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.205 ms | 0 - 17 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
45
+ | XLSR-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 0.393 ms | 0 - 17 MB | INT8 | NPU | Use Export Script |
46
+ | XLSR-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 0.911 ms | 0 - 18 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
47
+ | XLSR-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 2.332 ms | 0 - 15 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
48
+ | XLSR-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 1.206 ms | 0 - 11 MB | INT8 | NPU | Use Export Script |
49
+ | XLSR-Quantized | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 16.908 ms | 4 - 13 MB | INT8 | GPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
50
+ | XLSR-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.074 ms | 0 - 11 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
51
+ | XLSR-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 0.446 ms | 0 - 2 MB | INT8 | NPU | Use Export Script |
52
+ | XLSR-Quantized | SA7255P ADP | SA7255P | TFLITE | 4.013 ms | 2 - 12 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
53
+ | XLSR-Quantized | SA7255P ADP | SA7255P | QNN | 2.376 ms | 0 - 10 MB | INT8 | NPU | Use Export Script |
54
+ | XLSR-Quantized | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.055 ms | 0 - 11 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
55
+ | XLSR-Quantized | SA8255 (Proxy) | SA8255P Proxy | QNN | 0.444 ms | 0 - 3 MB | INT8 | NPU | Use Export Script |
56
+ | XLSR-Quantized | SA8295P ADP | SA8295P | TFLITE | 1.911 ms | 0 - 16 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
57
+ | XLSR-Quantized | SA8295P ADP | SA8295P | QNN | 1.059 ms | 0 - 14 MB | INT8 | NPU | Use Export Script |
58
+ | XLSR-Quantized | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.237 ms | 1 - 11 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
59
+ | XLSR-Quantized | SA8650 (Proxy) | SA8650P Proxy | QNN | 0.446 ms | 0 - 3 MB | INT8 | NPU | Use Export Script |
60
+ | XLSR-Quantized | SA8775P ADP | SA8775P | TFLITE | 2.083 ms | 0 - 11 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
61
+ | XLSR-Quantized | SA8775P ADP | SA8775P | QNN | 0.958 ms | 0 - 10 MB | INT8 | NPU | Use Export Script |
62
+ | XLSR-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 1.636 ms | 2 - 17 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
63
+ | XLSR-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 0.62 ms | 0 - 21 MB | INT8 | NPU | Use Export Script |
64
+ | XLSR-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.561 ms | 0 - 0 MB | INT8 | NPU | Use Export Script |
65
+ | XLSR-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.559 ms | 9 - 9 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
66
 
67
 
68
 
69
 
70
  ## Installation
71
 
 
72
 
73
+ Install the package via pip:
74
  ```bash
75
+ pip install "qai-hub-models[xlsr-quantized]"
76
  ```
77
 
78
 
 
79
  ## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device
80
 
81
  Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
 
127
  Device : Samsung Galaxy S23 (13)
128
  Runtime : TFLITE
129
  Estimated inference time (ms) : 1.1
130
+ Estimated peak memory usage (MB): [0, 10]
131
  Total # Ops : 19
132
  Compute Unit(s) : NPU (16 ops) CPU (3 ops)
133
  ```
 
154
  torch_model = Model.from_pretrained()
155
 
156
  # Device
157
+ device = hub.Device("Samsung Galaxy S24")
158
 
159
  # Trace model
160
  input_shape = torch_model.get_input_spec()
 
246
 
247
 
248
  ## License
249
+ * The license for the original implementation of XLSR-Quantized can be found
250
+ [here](https://github.com/quic/aimet-model-zoo/blob/develop/LICENSE.pdf).
251
  * The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)
252
 
253