v0.32.0

See https://github.com/quic/ai-hub-models/releases/v0.32.0 for changelog.

.gitattributes

@@ -37,3 +37,4 @@ CLIPTextEncoder.so filter=lfs diff=lfs merge=lfs -text
 CLIPImageEncoder.so filter=lfs diff=lfs merge=lfs -text
 OpenAI-Clip.so filter=lfs diff=lfs merge=lfs -text
 OpenAI-Clip.dlc filter=lfs diff=lfs merge=lfs -text
+DEPLOYMENT_MODEL_LICENSE.pdf filter=lfs diff=lfs merge=lfs -text

DEPLOYMENT_MODEL_LICENSE.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4409f93b0e82531303b3e10f52f1fdfb56467a25f05b7441c6bbd8bb8a64b42c
+size 109629

LICENSE

@@ -0,0 +1,2 @@
+The license of the original trained model can be found at https://github.com/openai/CLIP/blob/main/LICENSE.
+The license for the deployable model files (.tflite, .onnx, .dlc, .bin, etc.) can be found in DEPLOYMENT_MODEL_LICENSE.pdf.

OpenAI-Clip.onnx

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b4a07a75fc19e794b893cfcddb3ecac7c59b187815b7f9e193d5973b828222c7
-size 598785908
+oid sha256:17b702f6c2bbfcc19faf69cab19aad08e5eef7105ab22a29d10a73ede89f5cc7
+size 598804093

README.md

@@ -31,10 +31,8 @@ More details on model performance across various devices, can be found
   - Model checkpoint: ViT-B/16
   - Image input resolution: 224x224
   - Text context length: 77
-  - Number of parameters (CLIPTextEncoder): 76.0M
-  - Model size (CLIPTextEncoder): 290 MB
-  - Number of parameters (CLIPImageEncoder): 115M
-  - Model size (CLIPImageEncoder): 437 MB
+  - Number of parameters: 150M
+  - Model size (float): 571 MB
 
 | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
 |---|---|---|---|---|---|---|---|---|
@@ -58,15 +56,15 @@ More details on model performance across various devices, can be found
 | OpenAI-Clip | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 22.704 ms | 0 - 484 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
 | OpenAI-Clip | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 23.566 ms | 0 - 29 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
 | OpenAI-Clip | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN_DLC | 19.911 ms | 0 - 57 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
-| OpenAI-Clip | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 24.926 ms | 0 - 190 MB | NPU | [OpenAI-Clip.onnx](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx) |
+| OpenAI-Clip | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 24.694 ms | 0 - 132 MB | NPU | [OpenAI-Clip.onnx](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx) |
 | OpenAI-Clip | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 16.784 ms | 0 - 415 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
 | OpenAI-Clip | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_DLC | 14.146 ms | 1 - 491 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
-| OpenAI-Clip | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 17.72 ms | 1 - 535 MB | NPU | [OpenAI-Clip.onnx](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx) |
+| OpenAI-Clip | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 17.628 ms | 1 - 531 MB | NPU | [OpenAI-Clip.onnx](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx) |
 | OpenAI-Clip | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 13.737 ms | 0 - 406 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
 | OpenAI-Clip | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | QNN_DLC | 13.268 ms | 1 - 470 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
-| OpenAI-Clip | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 17.242 ms | 1 - 510 MB | NPU | [OpenAI-Clip.onnx](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx) |
+| OpenAI-Clip | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 16.816 ms | 1 - 513 MB | NPU | [OpenAI-Clip.onnx](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx) |
 | OpenAI-Clip | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 21.759 ms | 1667 - 1667 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
-| OpenAI-Clip | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 25.754 ms | 293 - 293 MB | NPU | [OpenAI-Clip.onnx](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx) |
+| OpenAI-Clip | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 25.714 ms | 293 - 293 MB | NPU | [OpenAI-Clip.onnx](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx) |