Yolo-R: Optimized for Qualcomm Devices

YoloR is a machine learning model that predicts bounding boxes and classes of objects in an image.

This is based on the implementation of Yolo-R 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

Due to licensing restrictions, we cannot distribute pre-exported model assets for this model. 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

See our repository for Yolo-R on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

• Input resolution: 640x640
• Model checkpoint: yolor_p6
• Model size (float): 17.9 MB
• Number of parameters: 4.68M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
Yolo-R	ONNX	float	Snapdragon® X2 Elite	27.222 ms	5 - 5 MB	NPU
Yolo-R	ONNX	float	Snapdragon® X Elite	45.796 ms	72 - 72 MB	NPU
Yolo-R	ONNX	float	Snapdragon® 8 Gen 3 Mobile	35.091 ms	1 - 249 MB	NPU
Yolo-R	ONNX	float	Snapdragon® 8 Gen 1 Mobile	72.384 ms	1 - 311 MB	NPU
Yolo-R	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	54.017 ms	5 - 13 MB	NPU
Yolo-R	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	45.68 ms	5 - 444 MB	NPU
Yolo-R	ONNX	float	Qualcomm® QCS8450	72.384 ms	1 - 311 MB	NPU
Yolo-R	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	51.648 ms	5 - 12 MB	NPU
Yolo-R	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	45.796 ms	72 - 72 MB	NPU
Yolo-R	ONNX	float	Qualcomm® Dragonwing™ Q-8750	30.266 ms	2 - 196 MB	NPU
Yolo-R	ONNX	float	Snapdragon® 8 Elite Mobile	30.266 ms	2 - 196 MB	NPU
Yolo-R	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	27.816 ms	3 - 202 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® X2 Elite	15.382 ms	2 - 2 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® X Elite	23.73 ms	37 - 37 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	15.901 ms	0 - 398 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	32.195 ms	3 - 405 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-8275	21.601 ms	1 - 7 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	22.789 ms	0 - 47 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® QCS8450	32.195 ms	3 - 405 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-9075	23.321 ms	1 - 6 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ IQ-X7181	23.73 ms	37 - 37 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ Q-7790	47.307 ms	3 - 320 MB	NPU
Yolo-R	ONNX	w8a16	Qualcomm® Dragonwing™ Q-8750	13.573 ms	1 - 334 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® 8 Elite Mobile	13.573 ms	1 - 334 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	14.619 ms	1 - 342 MB	NPU
Yolo-R	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	47.307 ms	3 - 320 MB	NPU
Yolo-R	QNN_DLC	float	Snapdragon® X2 Elite	18.205 ms	5 - 5 MB	NPU
Yolo-R	QNN_DLC	float	Snapdragon® X Elite	51.659 ms	5 - 5 MB	NPU
Yolo-R	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	40.865 ms	0 - 253 MB	NPU
Yolo-R	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	65.516 ms	5 - 312 MB	NPU
Yolo-R	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	58.207 ms	5 - 12 MB	NPU
Yolo-R	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	55.198 ms	5 - 406 MB	NPU
Yolo-R	QNN_DLC	float	Qualcomm® SA8775P	61.374 ms	1 - 197 MB	NPU
Yolo-R	QNN_DLC	float	Qualcomm® SA8650P	61.374 ms	1 - 197 MB	NPU
Yolo-R	QNN_DLC	float	Qualcomm® SA8255P	61.374 ms	1 - 197 MB	NPU
Yolo-R	QNN_DLC	float	Qualcomm® QCS8450	65.516 ms	5 - 312 MB	NPU
Yolo-R	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	61.449 ms	7 - 13 MB	NPU
Yolo-R	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	51.659 ms	5 - 5 MB	NPU
Yolo-R	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	22.853 ms	5 - 199 MB	NPU
Yolo-R	QNN_DLC	float	Qualcomm® SA7255P	124.404 ms	1 - 198 MB	NPU
Yolo-R	QNN_DLC	float	Snapdragon® 8 Elite Mobile	22.853 ms	5 - 199 MB	NPU
Yolo-R	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	19.092 ms	5 - 211 MB	NPU
Yolo-R	QNN_DLC	w8a16	Snapdragon® X2 Elite	7.053 ms	2 - 2 MB	NPU
Yolo-R	QNN_DLC	w8a16	Snapdragon® X Elite	16.333 ms	2 - 2 MB	NPU
Yolo-R	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	10.015 ms	2 - 382 MB	NPU
Yolo-R	QNN_DLC	w8a16	Snapdragon® 8 Gen 1 Mobile	25.198 ms	2 - 382 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS6490	65.097 ms	2 - 7 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-8275	13.954 ms	2 - 7 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® Dragonwing™ QCS8550 (Proxy)	15.338 ms	2 - 4 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® SA8775P	15.488 ms	1 - 310 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® SA8650P	15.488 ms	1 - 310 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® SA8255P	15.488 ms	1 - 310 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® QCS8450	25.198 ms	2 - 382 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-9075	15.889 ms	4 - 8 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® Dragonwing™ IQ-X7181	16.333 ms	2 - 2 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-6690	165.426 ms	2 - 350 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-7790	21.262 ms	2 - 326 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® Dragonwing™ Q-8750	7.358 ms	2 - 320 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® SA7255P	32.554 ms	1 - 311 MB	NPU
Yolo-R	QNN_DLC	w8a16	Qualcomm® SA8295P	20.795 ms	0 - 312 MB	NPU
Yolo-R	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	7.358 ms	2 - 320 MB	NPU
Yolo-R	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	5.854 ms	2 - 326 MB	NPU
Yolo-R	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	21.262 ms	2 - 326 MB	NPU

License

• The license for the original implementation of Yolo-R can be found here.

References

• You Only Learn One Representation: Unified Network for Multiple Tasks
• 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.