README.md

---
library_name: pytorch
license: other
tags:
- foundation
- android
pipeline_tag: image-classification

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/openai_clip/web-assets/model_demo.png)

# OpenAI-Clip: Optimized for Mobile Deployment
## Multi-modal foundational model for vision and language tasks like image/text similarity and for zero-shot image classification


Contrastive Language-Image Pre-Training (CLIP) uses a ViT like transformer to get visual features and a causal language model to get the text features. Both the text and visual features can then be used for a variety of zero-shot learning tasks.

This model is an implementation of OpenAI-Clip found [here](https://github.com/openai/CLIP/).


This repository provides scripts to run OpenAI-Clip on Qualcomm® devices.
More details on model performance across various devices, can be found
[here](https://aihub.qualcomm.com/models/openai_clip).



### Model Details

- **Model Type:** Model_use_case.image_classification
- **Model Stats:**
  - Model checkpoint: ViT-B/16
  - Image input resolution: 224x224
  - Text context length: 77
  - 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
|---|---|---|---|---|---|---|---|---|
| OpenAI-Clip | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 62.355 ms | 0 - 495 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
| OpenAI-Clip | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_DLC | 59.235 ms | 1 - 569 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
| OpenAI-Clip | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 24.652 ms | 0 - 499 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
| OpenAI-Clip | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_DLC | 25.491 ms | 0 - 527 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
| OpenAI-Clip | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 21.786 ms | 0 - 29 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
| OpenAI-Clip | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_DLC | 20.629 ms | 0 - 33 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
| OpenAI-Clip | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | ONNX | 22.4 ms | 1 - 37 MB | NPU | [OpenAI-Clip.onnx.zip](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx.zip) |
| OpenAI-Clip | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 24.838 ms | 0 - 495 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
| OpenAI-Clip | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_DLC | 108.559 ms | 1 - 569 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
| OpenAI-Clip | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 62.355 ms | 0 - 495 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
| OpenAI-Clip | float | SA7255P ADP | Qualcomm® SA7255P | QNN_DLC | 59.235 ms | 1 - 569 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
| OpenAI-Clip | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 21.79 ms | 0 - 35 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
| OpenAI-Clip | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_DLC | 20.713 ms | 0 - 39 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
| OpenAI-Clip | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 28.129 ms | 0 - 487 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
| OpenAI-Clip | float | SA8295P ADP | Qualcomm® SA8295P | QNN_DLC | 26.41 ms | 0 - 556 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
| OpenAI-Clip | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 21.69 ms | 0 - 27 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
| OpenAI-Clip | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_DLC | 20.779 ms | 0 - 28 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
| OpenAI-Clip | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 24.838 ms | 0 - 495 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
| OpenAI-Clip | float | SA8775P ADP | Qualcomm® SA8775P | QNN_DLC | 108.559 ms | 1 - 569 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 | TFLITE | 14.964 ms | 0 - 501 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 | 14.6 ms | 1 - 575 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 | 15.743 ms | 0 - 574 MB | NPU | [OpenAI-Clip.onnx.zip](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx.zip) |
| OpenAI-Clip | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | TFLITE | 11.878 ms | 0 - 496 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
| OpenAI-Clip | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_DLC | 11.455 ms | 0 - 551 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
| OpenAI-Clip | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | ONNX | 12.296 ms | 0 - 551 MB | NPU | [OpenAI-Clip.onnx.zip](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx.zip) |
| OpenAI-Clip | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | TFLITE | 9.267 ms | 0 - 493 MB | NPU | [OpenAI-Clip.tflite](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.tflite) |
| OpenAI-Clip | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_DLC | 9.309 ms | 0 - 496 MB | NPU | [OpenAI-Clip.dlc](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.dlc) |
| OpenAI-Clip | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | ONNX | 10.107 ms | 0 - 499 MB | NPU | [OpenAI-Clip.onnx.zip](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx.zip) |
| OpenAI-Clip | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_DLC | 21.684 ms | 1443 - 1443 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 | 22.723 ms | 295 - 295 MB | NPU | [OpenAI-Clip.onnx.zip](https://huggingface.co/qualcomm/OpenAI-Clip/blob/main/OpenAI-Clip.onnx.zip) |




## Installation


Install the package via pip:
```bash
pip install qai-hub-models
```


## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device

Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.aihub.qualcomm.com/) with your
Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.

With this API token, you can configure your client to run models on the cloud
hosted devices.
```bash
qai-hub configure --api_token API_TOKEN
```
Navigate to [docs](https://workbench.aihub.qualcomm.com/docs/) for more information.



## Demo off target

The package contains a simple end-to-end demo that downloads pre-trained
weights and runs this model on a sample input.

```bash
python -m qai_hub_models.models.openai_clip.demo
```

The above demo runs a reference implementation of pre-processing, model
inference, and post processing.

**NOTE**: If you want running in a Jupyter Notebook or Google Colab like
environment, please add the following to your cell (instead of the above).
```
%run -m qai_hub_models.models.openai_clip.demo
```


### Run model on a cloud-hosted device

In addition to the demo, you can also run the model on a cloud-hosted Qualcomm®
device. This script does the following:
* Performance check on-device on a cloud-hosted device
* Downloads compiled assets that can be deployed on-device for Android.
* Accuracy check between PyTorch and on-device outputs.

```bash
python -m qai_hub_models.models.openai_clip.export
```



## How does this work?

This [export script](https://aihub.qualcomm.com/models/openai_clip/qai_hub_models/models/OpenAI-Clip/export.py)
leverages [Qualcomm® AI Hub](https://aihub.qualcomm.com/) to optimize, validate, and deploy this model
on-device. Lets go through each step below in detail:

Step 1: **Compile model for on-device deployment**

To compile a PyTorch model for on-device deployment, we first trace the model
in memory using the `jit.trace` and then call the `submit_compile_job` API.

```python
import torch

import qai_hub as hub
from qai_hub_models.models.openai_clip import Model

# Load the model
torch_model = Model.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S25")

# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()

pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])

# Compile model on a specific device
compile_job = hub.submit_compile_job(
    model=pt_model,
    device=device,
    input_specs=torch_model.get_input_spec(),
)

# Get target model to run on-device
target_model = compile_job.get_target_model()

```


Step 2: **Performance profiling on cloud-hosted device**

After compiling models from step 1. Models can be profiled model on-device using the
`target_model`. Note that this scripts runs the model on a device automatically
provisioned in the cloud.  Once the job is submitted, you can navigate to a
provided job URL to view a variety of on-device performance metrics.
```python
profile_job = hub.submit_profile_job(
    model=target_model,
    device=device,
)
        
```

Step 3: **Verify on-device accuracy**

To verify the accuracy of the model on-device, you can run on-device inference
on sample input data on the same cloud hosted device.
```python
input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
    model=target_model,
    device=device,
    inputs=input_data,
)
    on_device_output = inference_job.download_output_data()

```
With the output of the model, you can compute like PSNR, relative errors or
spot check the output with expected output.

**Note**: This on-device profiling and inference requires access to Qualcomm®
AI Hub Workbench. [Sign up for access](https://myaccount.qualcomm.com/signup).




## Deploying compiled model to Android


The models can be deployed using multiple runtimes:
- TensorFlow Lite (`.tflite` export): [This
  tutorial](https://www.tensorflow.org/lite/android/quickstart) provides a
  guide to deploy the .tflite model in an Android application.


- QNN (`.so` export ): This [sample
  app](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-50/sample_app.html)
provides instructions on how to use the `.so` shared library  in an Android application.


## View on Qualcomm® AI Hub
Get more details on OpenAI-Clip's performance across various devices [here](https://aihub.qualcomm.com/models/openai_clip).
Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)


## License
* The license for the original implementation of OpenAI-Clip can be found
  [here](https://github.com/openai/CLIP/blob/main/LICENSE).
* 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)



## References
* [Learning Transferable Visual Models From Natural Language Supervision](https://arxiv.org/abs/2103.00020)
* [Source Model Implementation](https://github.com/openai/CLIP/)



## Community
* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
* For questions or feedback please [reach out to us](mailto:ai-hub-support@qti.qualcomm.com).