v0.33.1

See https://github.com/quic/ai-hub-models/releases/v0.33.1 for changelog.

README.md

@@ -34,31 +34,31 @@ More details on model performance across various devices, can be found
 
 | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
 |---|---|---|---|---|---|---|---|---|
-| AOT-GAN | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 621.515 ms | 0 - 142 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 621.953 ms | 0 - 142 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 548.03 ms | 1 - 214 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
-| AOT-GAN | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 202.975 ms | 3 - 157 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 203.353 ms | 3 - 157 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 225.613 ms | 4 - 179 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
-| AOT-GAN | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 152.482 ms | 3 - 28 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 156.157 ms | 3 - 28 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 123.459 ms | 3 - 53 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
-| AOT-GAN | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 203.677 ms | 3 - 145 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 203.608 ms | 3 - 145 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 166.762 ms | 1 - 216 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
-| AOT-GAN | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 621.515 ms | 0 - 142 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 621.953 ms | 0 - 142 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 548.03 ms | 1 - 214 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
-| AOT-GAN | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 153.831 ms | 3 - 29 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 156.143 ms | 2 - 28 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 122.831 ms | 3 - 52 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
-| AOT-GAN | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 226.919 ms | 3 - 135 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 226.885 ms | 0 - 132 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 187.301 ms | 1 - 186 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
-| AOT-GAN | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 156.422 ms | 3 - 27 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 153.62 ms | 3 - 29 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 122.021 ms | 3 - 43 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
-| AOT-GAN | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 203.677 ms | 3 - 145 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 203.608 ms | 3 - 145 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 166.762 ms | 1 - 216 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
-| AOT-GAN | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 156.036 ms | 3 - 26 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 152.989 ms | 3 - 28 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_DLC | 121.67 ms | 3 - 41 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
 | AOT-GAN | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 119.073 ms | 0 - 87 MB | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
-| AOT-GAN | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 114.438 ms | 1 - 164 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 114.224 ms | 3 - 166 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 89.241 ms | 4 - 217 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
 | AOT-GAN | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 87.108 ms | 6 - 211 MB | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
-| AOT-GAN | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 115.909 ms | 2 - 145 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
+| AOT-GAN | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 115.905 ms | 3 - 146 MB | NPU | [AOT-GAN.tflite](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.tflite) |
 | AOT-GAN | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_DLC | 84.29 ms | 4 - 294 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
 | AOT-GAN | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 81.125 ms | 7 - 285 MB | NPU | [AOT-GAN.onnx](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.onnx) |
 | AOT-GAN | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 148.402 ms | 79 - 79 MB | NPU | [AOT-GAN.dlc](https://huggingface.co/qualcomm/AOT-GAN/blob/main/AOT-GAN.dlc) |
@@ -126,7 +126,7 @@ Profiling Results
 AOT-GAN
 Device                          : cs_8275 (ANDROID 14)                 
 Runtime                         : TFLITE                               
-Estimated inference time (ms)   : 621.5                                
+Estimated inference time (ms)   : 622.0                                
 Estimated peak memory usage (MB): [0, 142]                             
 Total # Ops                     : 245                                  
 Compute Unit(s)                 : npu (245 ops) gpu (0 ops) cpu (0 ops)

precompiled/qualcomm-snapdragon-x-elite/AOT-GAN.onnx.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2bd66950b44df33201c8b142be559f755acc04db6497c8ce62c77ef8821d8cf
+oid sha256:dee2e3a1b5bf83604a2b154b47267e4a00cc368e3b5cef4d73a0e22a89a8844e
 size 31514960

precompiled/qualcomm-snapdragon-x-elite/sdk_versions.yml

@@ -1,5 +0,0 @@
-sdk_versions:
-  qnn_context_binary:
-    qairt: 2.34.2.250528164111_119506
-  precompiled_qnn_onnx:
-    qairt: 2.33.2.250410134701_117956