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	8.32 ms	3 - 3 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® X Elite	20.932 ms	174 - 174 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	14.312 ms	0 - 740 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	35.695 ms	4 - 762 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ QCS6490	128.395 ms	3 - 9 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	19.154 ms	1 - 8 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	20.312 ms	0 - 192 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® QCS8450	35.695 ms	4 - 762 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	24.091 ms	2 - 8 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	20.932 ms	174 - 174 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ Q-6690	214.248 ms	0 - 783 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	34.955 ms	6 - 582 MB	NPU
OWL-ViT	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	11.447 ms	0 - 663 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® 8 Elite Mobile	11.447 ms	0 - 663 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	7.735 ms	0 - 635 MB	NPU
OWL-ViT	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	34.955 ms	6 - 582 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® X2 Elite	19.285 ms	7 - 7 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® X Elite	45.544 ms	7 - 7 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	33.335 ms	4 - 659 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	63.06 ms	7 - 645 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	53.98 ms	7 - 16 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	135.024 ms	2 - 485 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	45.526 ms	7 - 810 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® SA8775P	52.015 ms	2 - 484 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® SA8650P	52.015 ms	2 - 484 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® SA8255P	52.015 ms	2 - 484 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® QCS8450	63.06 ms	7 - 645 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	51.802 ms	9 - 17 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	45.544 ms	7 - 7 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	26.973 ms	0 - 472 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® SA7255P	135.024 ms	2 - 485 MB	NPU
OWL-ViT	QNN_DLC	float	Qualcomm® SA8295P	60.616 ms	1 - 462 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® 8 Elite Mobile	26.973 ms	0 - 472 MB	NPU
OWL-ViT	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	20.271 ms	7 - 487 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Snapdragon® X2 Elite	23.468 ms	3 - 3 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Snapdragon® X Elite	86.383 ms	3 - 3 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	80.117 ms	4 - 989 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	94.872 ms	3 - 9 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	145.08 ms	1 - 749 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	96.977 ms	3 - 5 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® SA8775P	96.013 ms	0 - 747 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® SA8650P	96.013 ms	0 - 747 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® SA8255P	96.013 ms	0 - 747 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	93.754 ms	3 - 8 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	86.383 ms	3 - 3 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	68.137 ms	3 - 740 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Qualcomm® SA7255P	145.08 ms	1 - 749 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	68.137 ms	3 - 740 MB	NPU
OWL-ViT	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	21.851 ms	3 - 771 MB	NPU
OWL-ViT	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	23.339 ms	0 - 637 MB	NPU
OWL-ViT	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	48.576 ms	0 - 634 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	41.61 ms	0 - 311 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	116.202 ms	0 - 485 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	32.764 ms	0 - 4 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® SA8775P	39.287 ms	0 - 471 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® SA8650P	39.287 ms	0 - 471 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® SA8255P	39.287 ms	0 - 471 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® QCS8450	48.576 ms	0 - 634 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	44.952 ms	0 - 310 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	17.856 ms	0 - 479 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® SA7255P	116.202 ms	0 - 485 MB	NPU
OWL-ViT	TFLITE	float	Qualcomm® SA8295P	48.103 ms	0 - 460 MB	NPU
OWL-ViT	TFLITE	float	Snapdragon® 8 Elite Mobile	17.856 ms	0 - 479 MB	NPU
OWL-ViT	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	12.474 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.