qaihm-bot commited on
Commit
c429743
·
verified ·
1 Parent(s): 906b5f9

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +33 -33
README.md CHANGED
@@ -35,38 +35,38 @@ 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 | 1.088 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 | 0.447 ms | 0 - 2 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 | 1.445 ms | 0 - 8 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 | 0.901 ms | 0 - 23 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 | 0.303 ms | 0 - 19 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 | 1.07 ms | 0 - 21 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 | 1.182 ms | 0 - 18 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 | 0.471 ms | 0 - 17 MB | INT8 | NPU | Use Export Script |
46
- | XLSR-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 0.903 ms | 0 - 18 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
47
- | XLSR-Quantized | SA7255P ADP | SA7255P | TFLITE | 4.083 ms | 1 - 10 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
48
- | XLSR-Quantized | SA7255P ADP | SA7255P | QNN | 2.464 ms | 0 - 8 MB | INT8 | NPU | Use Export Script |
49
- | XLSR-Quantized | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.051 ms | 0 - 10 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
50
- | XLSR-Quantized | SA8255 (Proxy) | SA8255P Proxy | QNN | 0.45 ms | 0 - 2 MB | INT8 | NPU | Use Export Script |
51
- | XLSR-Quantized | SA8295P ADP | SA8295P | TFLITE | 1.938 ms | 0 - 12 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
52
- | XLSR-Quantized | SA8295P ADP | SA8295P | QNN | 1.503 ms | 0 - 13 MB | INT8 | NPU | Use Export Script |
53
- | XLSR-Quantized | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.072 ms | 0 - 11 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
54
- | XLSR-Quantized | SA8650 (Proxy) | SA8650P Proxy | QNN | 0.447 ms | 0 - 3 MB | INT8 | NPU | Use Export Script |
55
- | XLSR-Quantized | SA8775P ADP | SA8775P | TFLITE | 2.095 ms | 1 - 10 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
56
- | XLSR-Quantized | SA8775P ADP | SA8775P | QNN | 0.951 ms | 0 - 7 MB | INT8 | NPU | Use Export Script |
57
- | XLSR-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 2.274 ms | 0 - 15 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
58
- | XLSR-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 1.135 ms | 0 - 12 MB | INT8 | NPU | Use Export Script |
59
- | XLSR-Quantized | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 17.086 ms | 4 - 24 MB | INT8 | GPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
60
- | XLSR-Quantized | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 4.083 ms | 1 - 10 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
61
- | XLSR-Quantized | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 2.464 ms | 0 - 8 MB | INT8 | NPU | Use Export Script |
62
- | XLSR-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.071 ms | 0 - 4 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
63
- | XLSR-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 0.446 ms | 0 - 3 MB | INT8 | NPU | Use Export Script |
64
- | XLSR-Quantized | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 2.095 ms | 1 - 10 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
65
- | XLSR-Quantized | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 0.951 ms | 0 - 7 MB | INT8 | NPU | Use Export Script |
66
- | XLSR-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 1.691 ms | 1 - 17 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
67
- | XLSR-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 0.623 ms | 0 - 16 MB | INT8 | NPU | Use Export Script |
68
- | XLSR-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.542 ms | 0 - 0 MB | INT8 | NPU | Use Export Script |
69
- | XLSR-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.534 ms | 10 - 10 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
70
 
71
 
72
 
@@ -131,7 +131,7 @@ XLSR-Quantized
131
  Device : Samsung Galaxy S23 (13)
132
  Runtime : TFLITE
133
  Estimated inference time (ms) : 1.1
134
- Estimated peak memory usage (MB): [0, 10]
135
  Total # Ops : 19
136
  Compute Unit(s) : NPU (16 ops) CPU (3 ops)
137
  ```
 
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 | 1.058 ms | 0 - 17 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 | 0.422 ms | 0 - 2 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 | 1.349 ms | 0 - 17 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 | 0.879 ms | 0 - 22 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 | 0.246 ms | 0 - 20 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 | 0.972 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 | 1.145 ms | 0 - 18 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 | 0.306 ms | 0 - 18 MB | INT8 | NPU | Use Export Script |
46
+ | XLSR-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 0.874 ms | 0 - 18 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
47
+ | XLSR-Quantized | SA7255P ADP | SA7255P | TFLITE | 4.027 ms | 2 - 12 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
48
+ | XLSR-Quantized | SA7255P ADP | SA7255P | QNN | 2.108 ms | 0 - 8 MB | INT8 | NPU | Use Export Script |
49
+ | XLSR-Quantized | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 1.067 ms | 0 - 17 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
50
+ | XLSR-Quantized | SA8255 (Proxy) | SA8255P Proxy | QNN | 0.389 ms | 0 - 2 MB | INT8 | NPU | Use Export Script |
51
+ | XLSR-Quantized | SA8295P ADP | SA8295P | TFLITE | 1.842 ms | 0 - 16 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
52
+ | XLSR-Quantized | SA8295P ADP | SA8295P | QNN | 0.737 ms | 0 - 14 MB | INT8 | NPU | Use Export Script |
53
+ | XLSR-Quantized | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 1.072 ms | 0 - 17 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
54
+ | XLSR-Quantized | SA8650 (Proxy) | SA8650P Proxy | QNN | 0.412 ms | 0 - 3 MB | INT8 | NPU | Use Export Script |
55
+ | XLSR-Quantized | SA8775P ADP | SA8775P | TFLITE | 2.06 ms | 0 - 12 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
56
+ | XLSR-Quantized | SA8775P ADP | SA8775P | QNN | 0.71 ms | 0 - 10 MB | INT8 | NPU | Use Export Script |
57
+ | XLSR-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | TFLITE | 2.209 ms | 0 - 15 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
58
+ | XLSR-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 1.035 ms | 0 - 11 MB | INT8 | NPU | Use Export Script |
59
+ | XLSR-Quantized | RB5 (Proxy) | QCS8250 Proxy | TFLITE | 16.831 ms | 4 - 13 MB | INT8 | GPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
60
+ | XLSR-Quantized | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 4.027 ms | 2 - 12 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
61
+ | XLSR-Quantized | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 2.108 ms | 0 - 8 MB | INT8 | NPU | Use Export Script |
62
+ | XLSR-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 1.098 ms | 0 - 17 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
63
+ | XLSR-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 0.391 ms | 0 - 2 MB | INT8 | NPU | Use Export Script |
64
+ | XLSR-Quantized | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 2.06 ms | 0 - 12 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
65
+ | XLSR-Quantized | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 0.71 ms | 0 - 10 MB | INT8 | NPU | Use Export Script |
66
+ | XLSR-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 1.82 ms | 2 - 19 MB | INT8 | NPU | [XLSR-Quantized.tflite](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.tflite) |
67
+ | XLSR-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 0.478 ms | 0 - 20 MB | INT8 | NPU | Use Export Script |
68
+ | XLSR-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 0.484 ms | 0 - 0 MB | INT8 | NPU | Use Export Script |
69
+ | XLSR-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 1.379 ms | 9 - 9 MB | INT8 | NPU | [XLSR-Quantized.onnx](https://huggingface.co/qualcomm/XLSR-Quantized/blob/main/XLSR-Quantized.onnx) |
70
 
71
 
72
 
 
131
  Device : Samsung Galaxy S23 (13)
132
  Runtime : TFLITE
133
  Estimated inference time (ms) : 1.1
134
+ Estimated peak memory usage (MB): [0, 17]
135
  Total # Ops : 19
136
  Compute Unit(s) : NPU (16 ops) CPU (3 ops)
137
  ```