OWL-ViT: Optimized for Qualcomm Devices

OWL-ViT (Open-World Localization with Vision Transformers) is an open-vocabulary object detector that uses a CLIP-based ViT-B/32 backbone. Given an image and one or more free-form text queries, the model predicts bounding boxes and confidence scores for each query.

This is based on the implementation of OWL-ViT 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	w8a16	Universal	QAIRT 2.45, ONNX Runtime 1.27.1	Download
QNN_DLC	float	Universal	QAIRT 2.45	Download
QNN_DLC	w8a16	Universal	QAIRT 2.45	Download
TFLITE	float	Universal	QAIRT 2.45	Download

For more device-specific assets and performance metrics, visit OWL-ViT 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 OWL-ViT on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

• Input image resolution: 768x768
• Model checkpoint: google/owlvit-base-patch32
• Model size (float): 613 MB
• Number of parameters: 153M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
OWL-ViT	ONNX	w8a16	Snapdragon® X2 Elite	8.296 ms	3 - 3 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® X Elite	21.123 ms	175 - 175 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	14.343 ms	4 - 743 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	36.127 ms	4 - 761 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ QCS6490	128.224 ms	3 - 9 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	19.44 ms	1 - 8 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	20.459 ms	0 - 192 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® QCS8450	36.127 ms	4 - 761 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	21.068 ms	0 - 6 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	21.123 ms	175 - 175 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	231.005 ms	2 - 784 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	34.748 ms	6 - 581 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	11.453 ms	0 - 664 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® 8 Elite Mobile	11.453 ms	0 - 664 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	7.727 ms	2 - 636 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	34.748 ms	6 - 581 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® X2 Elite	19.555 ms	7 - 7 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® X Elite	45.652 ms	7 - 7 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	33.349 ms	6 - 662 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	62.768 ms	7 - 644 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	54.136 ms	7 - 16 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	45.604 ms	7 - 9 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® SA8775P	52.042 ms	2 - 483 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® SA8650P	52.042 ms	2 - 483 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® SA8255P	52.042 ms	2 - 483 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® QCS8450	62.768 ms	7 - 644 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	51.669 ms	9 - 17 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	45.652 ms	7 - 7 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	27.167 ms	7 - 479 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® SA7255P	134.99 ms	2 - 484 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® SA8295P	60.525 ms	0 - 461 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® 8 Elite Mobile	27.167 ms	7 - 479 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	20.235 ms	0 - 481 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Snapdragon® X2 Elite	23.399 ms	3 - 3 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Snapdragon® X Elite	86.371 ms	3 - 3 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	80.096 ms	3 - 987 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	94.846 ms	3 - 9 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	99.17 ms	4 - 6 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® SA8775P	95.803 ms	0 - 747 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® SA8650P	95.803 ms	0 - 747 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® SA8255P	95.803 ms	0 - 747 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	93.714 ms	3 - 8 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	86.371 ms	3 - 3 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	68.365 ms	3 - 741 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	68.365 ms	3 - 741 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	21.858 ms	3 - 771 MB	NPU
OWL-ViT	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	23.495 ms	0 - 640 MB	NPU
OWL-ViT	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	48.886 ms	0 - 632 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	41.582 ms	0 - 311 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	32.552 ms	0 - 3 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® SA8775P	39.316 ms	0 - 471 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® SA8650P	39.316 ms	0 - 471 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® SA8255P	39.316 ms	0 - 471 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® QCS8450	48.886 ms	0 - 632 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	39.593 ms	0 - 310 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	17.672 ms	0 - 473 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® SA7255P	116.226 ms	0 - 484 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® SA8295P	47.986 ms	0 - 461 MB	NPU
OWL-ViT	TFLITE	float	Snapdragon® 8 Elite Mobile	17.672 ms	0 - 473 MB	NPU
OWL-ViT	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	12.452 ms	0 - 496 MB	NPU

License

• The license for the original implementation of OWL-ViT can be found here.

References

• Simple Open-Vocabulary Object Detection with Vision Transformers
• 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.