Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -36,17 +36,18 @@ More details on model performance across various devices, can be found
|
|
36 |
|
37 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
38 |
|---|---|---|---|---|---|---|---|---|
|
39 |
-
| ConvNext-Tiny-w8a16-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 3.
|
40 |
-
| ConvNext-Tiny-w8a16-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 2.
|
41 |
-
| ConvNext-Tiny-w8a16-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 2.
|
42 |
-
| ConvNext-Tiny-w8a16-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 13.
|
43 |
-
| ConvNext-Tiny-w8a16-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 3.
|
44 |
-
| ConvNext-Tiny-w8a16-Quantized |
|
45 |
-
| ConvNext-Tiny-w8a16-Quantized |
|
46 |
-
| ConvNext-Tiny-w8a16-Quantized |
|
47 |
-
| ConvNext-Tiny-w8a16-Quantized |
|
48 |
-
| ConvNext-Tiny-w8a16-Quantized |
|
49 |
-
| ConvNext-Tiny-w8a16-Quantized |
|
|
|
50 |
|
51 |
|
52 |
|
@@ -111,8 +112,8 @@ Profiling Results
|
|
111 |
ConvNext-Tiny-w8a16-Quantized
|
112 |
Device : Samsung Galaxy S23 (13)
|
113 |
Runtime : QNN
|
114 |
-
Estimated inference time (ms) : 3.
|
115 |
-
Estimated peak memory usage (MB): [0,
|
116 |
Total # Ops : 215
|
117 |
Compute Unit(s) : NPU (215 ops)
|
118 |
```
|
|
|
36 |
|
37 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
38 |
|---|---|---|---|---|---|---|---|---|
|
39 |
+
| ConvNext-Tiny-w8a16-Quantized | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 3.449 ms | 0 - 86 MB | INT8 | NPU | [ConvNext-Tiny-w8a16-Quantized.so](https://huggingface.co/qualcomm/ConvNext-Tiny-w8a16-Quantized/blob/main/ConvNext-Tiny-w8a16-Quantized.so) |
|
40 |
+
| ConvNext-Tiny-w8a16-Quantized | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 2.425 ms | 1 - 35 MB | INT8 | NPU | [ConvNext-Tiny-w8a16-Quantized.so](https://huggingface.co/qualcomm/ConvNext-Tiny-w8a16-Quantized/blob/main/ConvNext-Tiny-w8a16-Quantized.so) |
|
41 |
+
| ConvNext-Tiny-w8a16-Quantized | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 2.444 ms | 0 - 36 MB | INT8 | NPU | Use Export Script |
|
42 |
+
| ConvNext-Tiny-w8a16-Quantized | RB3 Gen 2 (Proxy) | QCS6490 Proxy | QNN | 13.636 ms | 1 - 8 MB | INT8 | NPU | Use Export Script |
|
43 |
+
| ConvNext-Tiny-w8a16-Quantized | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 3.1 ms | 0 - 2 MB | INT8 | NPU | Use Export Script |
|
44 |
+
| ConvNext-Tiny-w8a16-Quantized | SA7255P ADP | SA7255P | QNN | 26.853 ms | 1 - 6 MB | INT8 | NPU | Use Export Script |
|
45 |
+
| ConvNext-Tiny-w8a16-Quantized | SA8255 (Proxy) | SA8255P Proxy | QNN | 3.11 ms | 0 - 2 MB | INT8 | NPU | Use Export Script |
|
46 |
+
| ConvNext-Tiny-w8a16-Quantized | SA8295P ADP | SA8295P | QNN | 4.817 ms | 0 - 6 MB | INT8 | NPU | Use Export Script |
|
47 |
+
| ConvNext-Tiny-w8a16-Quantized | SA8650 (Proxy) | SA8650P Proxy | QNN | 3.107 ms | 0 - 1 MB | INT8 | NPU | Use Export Script |
|
48 |
+
| ConvNext-Tiny-w8a16-Quantized | SA8775P ADP | SA8775P | QNN | 4.469 ms | 0 - 6 MB | INT8 | NPU | Use Export Script |
|
49 |
+
| ConvNext-Tiny-w8a16-Quantized | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 4.227 ms | 0 - 40 MB | INT8 | NPU | Use Export Script |
|
50 |
+
| ConvNext-Tiny-w8a16-Quantized | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 3.383 ms | 0 - 0 MB | INT8 | NPU | Use Export Script |
|
51 |
|
52 |
|
53 |
|
|
|
112 |
ConvNext-Tiny-w8a16-Quantized
|
113 |
Device : Samsung Galaxy S23 (13)
|
114 |
Runtime : QNN
|
115 |
+
Estimated inference time (ms) : 3.4
|
116 |
+
Estimated peak memory usage (MB): [0, 86]
|
117 |
Total # Ops : 215
|
118 |
Compute Unit(s) : NPU (215 ops)
|
119 |
```
|