Hugging Face's logo

qualcomm
/
OpenAI-Clip

Image Classification
PyTorch
TF Lite
foundation
android
Model card Files Community
1
qaihm-bot commited on
Commit
1b969af
•
1 Parent(s): 41e676d

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +19 -19
README.md CHANGED
@@ -36,10 +36,10 @@ More details on model performance across various devices, can be found
36
 
37
  | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
38
  | ---|---|---|---|---|---|---|---|
39
- | Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 15.528 ms | 0 - 3 MB | FP16 | NPU | [CLIPTextEncoder.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/CLIPTextEncoder.tflite)
40
- | Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 127.729 ms | 0 - 4 MB | FP16 | NPU | [CLIPImageEncoder.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/CLIPImageEncoder.tflite)
41
- | Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 8.149 ms | 0 - 23 MB | FP16 | NPU | [CLIPTextEncoder.so](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/CLIPTextEncoder.so)
42
- | Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 50.903 ms | 0 - 57 MB | FP16 | NPU | [CLIPImageEncoder.so](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/CLIPImageEncoder.so)
43
 
44
 
45
  ## Installation
@@ -100,31 +100,31 @@ python -m qai_hub_models.models.openai_clip.export
100
  ```
101
  Profile Job summary of CLIPTextEncoder
102
  --------------------------------------------------
103
- Device: Samsung Galaxy S23 Ultra (13)
104
- Estimated Inference Time: 15.53 ms
105
- Estimated Peak Memory Range: 0.04-2.96 MB
106
  Compute Units: NPU (574),CPU (2) | Total (576)
107
 
108
  Profile Job summary of CLIPImageEncoder
109
  --------------------------------------------------
110
- Device: Samsung Galaxy S23 Ultra (13)
111
- Estimated Inference Time: 127.73 ms
112
- Estimated Peak Memory Range: 0.15-3.69 MB
113
  Compute Units: NPU (575) | Total (575)
114
 
115
  Profile Job summary of CLIPTextEncoder
116
  --------------------------------------------------
117
- Device: Samsung Galaxy S23 Ultra (13)
118
- Estimated Inference Time: 8.15 ms
119
- Estimated Peak Memory Range: 0.04-22.63 MB
120
- Compute Units: NPU (377) | Total (377)
121
 
122
  Profile Job summary of CLIPImageEncoder
123
  --------------------------------------------------
124
- Device: Samsung Galaxy S23 Ultra (13)
125
- Estimated Inference Time: 50.90 ms
126
- Estimated Peak Memory Range: 0.08-56.97 MB
127
- Compute Units: NPU (370) | Total (370)
128
 
129
 
130
  ```
@@ -229,7 +229,7 @@ Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
229
  ## License
230
  - The license for the original implementation of OpenAI-Clip can be found
231
  [here](https://github.com/openai/CLIP/blob/main/LICENSE).
232
- - 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).
233
 
234
  ## References
235
  * [Learning Transferable Visual Models From Natural Language Supervision](https://arxiv.org/abs/2103.00020)
 
36
 
37
  | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
38
  | ---|---|---|---|---|---|---|---|
39
+ | Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 15.516 ms | 0 - 3 MB | FP16 | NPU | [CLIPTextEncoder.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/CLIPTextEncoder.tflite)
40
+ | Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | TFLite | 128.196 ms | 0 - 4 MB | FP16 | NPU | [CLIPImageEncoder.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/CLIPImageEncoder.tflite)
41
+ | Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 15.586 ms | 0 - 3 MB | FP16 | NPU | [CLIPTextEncoder.so](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/CLIPTextEncoder.so)
42
+ | Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 127.795 ms | 0 - 4 MB | FP16 | NPU | [CLIPImageEncoder.so](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/CLIPImageEncoder.so)
43
 
44
 
45
  ## Installation
 
100
  ```
101
  Profile Job summary of CLIPTextEncoder
102
  --------------------------------------------------
103
+ Device: Samsung Galaxy S24 (14)
104
+ Estimated Inference Time: 11.12 ms
105
+ Estimated Peak Memory Range: 0.02-194.85 MB
106
  Compute Units: NPU (574),CPU (2) | Total (576)
107
 
108
  Profile Job summary of CLIPImageEncoder
109
  --------------------------------------------------
110
+ Device: Samsung Galaxy S24 (14)
111
+ Estimated Inference Time: 98.56 ms
112
+ Estimated Peak Memory Range: 0.16-745.19 MB
113
  Compute Units: NPU (575) | Total (575)
114
 
115
  Profile Job summary of CLIPTextEncoder
116
  --------------------------------------------------
117
+ Device: Samsung Galaxy S24 (14)
118
+ Estimated Inference Time: 11.25 ms
119
+ Estimated Peak Memory Range: 0.04-195.98 MB
120
+ Compute Units: NPU (574),CPU (2) | Total (576)
121
 
122
  Profile Job summary of CLIPImageEncoder
123
  --------------------------------------------------
124
+ Device: Samsung Galaxy S24 (14)
125
+ Estimated Inference Time: 97.28 ms
126
+ Estimated Peak Memory Range: 0.23-747.56 MB
127
+ Compute Units: NPU (575) | Total (575)
128
 
129
 
130
  ```
 
229
  ## License
230
  - The license for the original implementation of OpenAI-Clip can be found
231
  [here](https://github.com/openai/CLIP/blob/main/LICENSE).
232
+ - The license for the compiled assets for on-device deployment can be found [here]({deploy_license_url})
233
 
234
  ## References
235
  * [Learning Transferable Visual Models From Natural Language Supervision](https://arxiv.org/abs/2103.00020)