YOLOv10-Detection: Optimized for Qualcomm Devices

Ultralytics YOLOv10 is a machine learning model that predicts bounding boxes and classes of objects in an image.

This is based on the implementation of YOLOv10-Detection 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 YOLOv10-Detection on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

• Model checkpoint: YOLOv10-N
• Input resolution: 640x640
• Number of parameters: 2.33M
• Model size (float): 8.95 MB
• Model size (w8a8): 2.55 MB
• Model size (w8a16): 3.04 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
YOLOv10-Detection	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	2.724 ms	1 - 220 MB	NPU
YOLOv10-Detection	ONNX	float	Snapdragon® X2 Elite	3.092 ms	3 - 3 MB	NPU
YOLOv10-Detection	ONNX	float	Snapdragon® X Elite	6.641 ms	5 - 5 MB	NPU
YOLOv10-Detection	ONNX	float	Snapdragon® 8 Gen 3 Mobile	3.827 ms	0 - 273 MB	NPU
YOLOv10-Detection	ONNX	float	Qualcomm® QCS8550 (Proxy)	6.129 ms	5 - 9 MB	NPU
YOLOv10-Detection	ONNX	float	Qualcomm® QCS9075	6.995 ms	5 - 7 MB	NPU
YOLOv10-Detection	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	3.065 ms	0 - 223 MB	NPU
YOLOv10-Detection	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	2.101 ms	0 - 80 MB	NPU
YOLOv10-Detection	ONNX	w8a16	Snapdragon® X2 Elite	2.241 ms	0 - 0 MB	NPU
YOLOv10-Detection	ONNX	w8a16	Snapdragon® X Elite	5.577 ms	2 - 2 MB	NPU
YOLOv10-Detection	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	3.159 ms	0 - 105 MB	NPU
YOLOv10-Detection	ONNX	w8a16	Qualcomm® QCS6490	334.821 ms	66 - 71 MB	CPU
YOLOv10-Detection	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	5.08 ms	2 - 6 MB	NPU
YOLOv10-Detection	ONNX	w8a16	Qualcomm® QCS9075	5.896 ms	2 - 5 MB	NPU
YOLOv10-Detection	ONNX	w8a16	Qualcomm® QCM6690	168.921 ms	57 - 66 MB	CPU
YOLOv10-Detection	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	2.302 ms	0 - 79 MB	NPU
YOLOv10-Detection	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	149.95 ms	70 - 79 MB	CPU
YOLOv10-Detection	ONNX	w8a8_mixed_int16	Snapdragon® 8 Elite Gen 5 Mobile	6.216 ms	8 - 320 MB	NPU
YOLOv10-Detection	ONNX	w8a8_mixed_int16	Snapdragon® X2 Elite	6.664 ms	10 - 10 MB	NPU
YOLOv10-Detection	ONNX	w8a8_mixed_int16	Snapdragon® X Elite	14.842 ms	8 - 8 MB	NPU
YOLOv10-Detection	ONNX	w8a8_mixed_int16	Snapdragon® 8 Gen 3 Mobile	8.453 ms	8 - 239 MB	NPU
YOLOv10-Detection	ONNX	w8a8_mixed_int16	Qualcomm® QCS6490	142.685 ms	58 - 69 MB	CPU
YOLOv10-Detection	ONNX	w8a8_mixed_int16	Qualcomm® QCS8550 (Proxy)	11.741 ms	5 - 43 MB	NPU
YOLOv10-Detection	ONNX	w8a8_mixed_int16	Qualcomm® QCS9075	16.499 ms	7 - 10 MB	NPU
YOLOv10-Detection	ONNX	w8a8_mixed_int16	Qualcomm® QCM6690	79.556 ms	53 - 64 MB	CPU
YOLOv10-Detection	ONNX	w8a8_mixed_int16	Snapdragon® 8 Elite For Galaxy Mobile	6.378 ms	6 - 192 MB	NPU
YOLOv10-Detection	ONNX	w8a8_mixed_int16	Snapdragon® 7 Gen 4 Mobile	71.336 ms	69 - 80 MB	CPU
YOLOv10-Detection	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	1.869 ms	0 - 185 MB	NPU
YOLOv10-Detection	QNN_DLC	float	Snapdragon® X2 Elite	2.559 ms	5 - 5 MB	NPU
YOLOv10-Detection	QNN_DLC	float	Snapdragon® X Elite	4.367 ms	5 - 5 MB	NPU
YOLOv10-Detection	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	2.952 ms	0 - 211 MB	NPU
YOLOv10-Detection	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	3.984 ms	5 - 7 MB	NPU
YOLOv10-Detection	QNN_DLC	float	Qualcomm® SA8775P	5.442 ms	0 - 184 MB	NPU
YOLOv10-Detection	QNN_DLC	float	Qualcomm® QCS9075	5.129 ms	5 - 11 MB	NPU
YOLOv10-Detection	QNN_DLC	float	Qualcomm® QCS8450 (Proxy)	9.199 ms	3 - 200 MB	NPU
YOLOv10-Detection	QNN_DLC	float	Qualcomm® SA8295P	8.881 ms	0 - 167 MB	NPU
YOLOv10-Detection	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	2.199 ms	0 - 186 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	1.767 ms	2 - 184 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Snapdragon® X2 Elite	2.188 ms	2 - 2 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Snapdragon® X Elite	4.664 ms	2 - 2 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	3.022 ms	2 - 202 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Qualcomm® QCS8275 (Proxy)	8.182 ms	2 - 176 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	4.29 ms	0 - 5 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Qualcomm® SA8775P	4.919 ms	0 - 177 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Qualcomm® QCS9075	4.922 ms	0 - 4 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Qualcomm® QCM6690	18.813 ms	2 - 183 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Qualcomm® SA7255P	8.182 ms	2 - 176 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	2.109 ms	2 - 180 MB	NPU
YOLOv10-Detection	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	4.52 ms	2 - 180 MB	NPU
YOLOv10-Detection	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	1.579 ms	0 - 91 MB	NPU
YOLOv10-Detection	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	2.647 ms	0 - 116 MB	NPU
YOLOv10-Detection	TFLITE	float	Qualcomm® QCS8275 (Proxy)	12.57 ms	0 - 81 MB	NPU
YOLOv10-Detection	TFLITE	float	Qualcomm® QCS8550 (Proxy)	3.578 ms	0 - 2 MB	NPU
YOLOv10-Detection	TFLITE	float	Qualcomm® SA8775P	5.102 ms	0 - 84 MB	NPU
YOLOv10-Detection	TFLITE	float	Qualcomm® QCS9075	4.817 ms	0 - 13 MB	NPU
YOLOv10-Detection	TFLITE	float	Qualcomm® QCS8450 (Proxy)	7.707 ms	0 - 197 MB	NPU
YOLOv10-Detection	TFLITE	float	Qualcomm® SA7255P	12.57 ms	0 - 81 MB	NPU
YOLOv10-Detection	TFLITE	float	Qualcomm® SA8295P	8.136 ms	0 - 170 MB	NPU
YOLOv10-Detection	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	1.99 ms	0 - 83 MB	NPU

License

• The license for the original implementation of YOLOv10-Detection can be found here.

References

• YOLOv10: Real-Time End-to-End Object Detection
• 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.