OpenAI-Clip: Optimized for Qualcomm Devices

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 is based on the implementation of OpenAI-Clip found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Getting Started

There are two ways to deploy this model on your device:

Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

Runtime	Precision	Chipset	SDK Versions	Download
ONNX	float	Universal	QAIRT 2.42, ONNX Runtime 1.24.3	Download
QNN_DLC	float	Universal	QAIRT 2.43	Download
TFLITE	float	Universal	QAIRT 2.43, TFLite 2.19.1	Download

For more device-specific assets and performance metrics, visit OpenAI-Clip on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:

• Custom weights (e.g., fine-tuned checkpoints)
• Custom input shapes
• Target device and runtime configurations

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for OpenAI-Clip on GitHub for usage instructions.

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

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
OpenAI-Clip	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	7.001 ms	1 - 497 MB	NPU
OpenAI-Clip	ONNX	float	Snapdragon® X2 Elite	7.197 ms	291 - 291 MB	NPU
OpenAI-Clip	ONNX	float	Snapdragon® X Elite	16.459 ms	291 - 291 MB	NPU
OpenAI-Clip	ONNX	float	Snapdragon® 8 Gen 3 Mobile	11.243 ms	1 - 573 MB	NPU
OpenAI-Clip	ONNX	float	Qualcomm® QCS8550 (Proxy)	15.75 ms	0 - 322 MB	NPU
OpenAI-Clip	ONNX	float	Qualcomm® QCS9075	20.371 ms	0 - 4 MB	NPU
OpenAI-Clip	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	9.097 ms	0 - 500 MB	NPU
OpenAI-Clip	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	8.343 ms	0 - 486 MB	NPU
OpenAI-Clip	QNN_DLC	float	Snapdragon® X2 Elite	9.094 ms	1 - 1 MB	NPU
OpenAI-Clip	QNN_DLC	float	Snapdragon® X Elite	18.809 ms	1 - 1 MB	NPU
OpenAI-Clip	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	12.623 ms	0 - 550 MB	NPU
OpenAI-Clip	QNN_DLC	float	Qualcomm® QCS8275 (Proxy)	56.026 ms	1 - 506 MB	NPU
OpenAI-Clip	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	17.919 ms	1 - 3 MB	NPU
OpenAI-Clip	QNN_DLC	float	Qualcomm® SA8775P	20.929 ms	1 - 504 MB	NPU
OpenAI-Clip	QNN_DLC	float	Qualcomm® QCS9075	20.981 ms	1 - 3 MB	NPU
OpenAI-Clip	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	20.901 ms	0 - 502 MB	NPU
OpenAI-Clip	QNN_DLC	float	Qualcomm® SA7255P	56.026 ms	1 - 506 MB	NPU
OpenAI-Clip	QNN_DLC	float	Qualcomm® SA8295P	22.076 ms	0 - 496 MB	NPU
OpenAI-Clip	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	10.458 ms	1 - 514 MB	NPU
OpenAI-Clip	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	8.351 ms	0 - 501 MB	NPU
OpenAI-Clip	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	12.82 ms	0 - 566 MB	NPU
OpenAI-Clip	TFLITE	float	Qualcomm® QCS8275 (Proxy)	56.078 ms	0 - 513 MB	NPU
OpenAI-Clip	TFLITE	float	Qualcomm® QCS8550 (Proxy)	17.476 ms	0 - 3 MB	NPU
OpenAI-Clip	TFLITE	float	Qualcomm® SA8775P	20.942 ms	0 - 513 MB	NPU
OpenAI-Clip	TFLITE	float	Qualcomm® QCS9075	20.471 ms	0 - 294 MB	NPU
OpenAI-Clip	TFLITE	float	Qualcomm® QCS8450 (Proxy)	21.069 ms	0 - 507 MB	NPU
OpenAI-Clip	TFLITE	float	Qualcomm® SA7255P	56.078 ms	0 - 513 MB	NPU
OpenAI-Clip	TFLITE	float	Qualcomm® SA8295P	21.947 ms	0 - 500 MB	NPU
OpenAI-Clip	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	10.605 ms	0 - 529 MB	NPU

License

• The license for the original implementation of OpenAI-Clip can be found here.

References

• Learning Transferable Visual Models From Natural Language Supervision
• Source Model Implementation

Community

• Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
• For questions or feedback please reach out to us.