YOLO26-Pose: Optimized for Qualcomm Devices

Ultralytics YOLO26-Pose is a machine learning model that predicts human body keypoints and bounding boxes in an image.

This is based on the implementation of YOLO26-Pose 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 YOLO26-Pose on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.pose_estimation

Model Stats:

• Model checkpoint: YOLO26-N-Pose
• Input resolution: 640x640
• Number of parameters: 2.97M
• Model size (float): 11.4 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
YOLO26-Pose	ONNX	float	Snapdragon® X2 Elite	4.745 ms	208 - 208 MB	NPU
YOLO26-Pose	ONNX	float	Snapdragon® X Elite	10.151 ms	145 - 145 MB	NPU
YOLO26-Pose	ONNX	float	Snapdragon® 8 Gen 3 Mobile	6.894 ms	9 - 204 MB	NPU
YOLO26-Pose	ONNX	float	Snapdragon® 8 Gen 1 Mobile	18.053 ms	1 - 212 MB	NPU
YOLO26-Pose	ONNX	float	Qualcomm® QCS8550 (Proxy)	9.892 ms	6 - 10 MB	NPU
YOLO26-Pose	ONNX	float	Qualcomm® QCS8450	18.053 ms	1 - 212 MB	NPU
YOLO26-Pose	ONNX	float	Snapdragon® 8 Elite Mobile	4.763 ms	6 - 187 MB	NPU
YOLO26-Pose	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	4.362 ms	5 - 184 MB	NPU
YOLO26-Pose	ONNX	float	Qualcomm® QCS9075	15.466 ms	7 - 53 MB	NPU
YOLO26-Pose	ONNX	float	Qualcomm® QCS8750	4.763 ms	6 - 187 MB	NPU
YOLO26-Pose	ONNX	float	Qualcomm® QCS7181	10.151 ms	145 - 145 MB	NPU
YOLO26-Pose	ONNX	w8a16	Snapdragon® X2 Elite	2.633 ms	210 - 210 MB	NPU
YOLO26-Pose	ONNX	w8a16	Snapdragon® X Elite	6.658 ms	179 - 179 MB	NPU
YOLO26-Pose	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	3.838 ms	0 - 223 MB	NPU
YOLO26-Pose	ONNX	w8a16	Snapdragon® 8 Gen 1 Mobile	7.955 ms	4 - 225 MB	NPU
YOLO26-Pose	ONNX	w8a16	Qualcomm® QCS6490	19.36 ms	1 - 48 MB	NPU
YOLO26-Pose	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	6.287 ms	2 - 15 MB	NPU
YOLO26-Pose	ONNX	w8a16	Qualcomm® QCS8450	7.955 ms	4 - 225 MB	NPU
YOLO26-Pose	ONNX	w8a16	Qualcomm® QCS9075	6.9 ms	2 - 48 MB	NPU
YOLO26-Pose	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	2.472 ms	0 - 90 MB	NPU
YOLO26-Pose	ONNX	w8a16	Snapdragon® 8 Elite Mobile	2.821 ms	0 - 93 MB	NPU
YOLO26-Pose	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	10.248 ms	4 - 213 MB	NPU
YOLO26-Pose	ONNX	w8a16	Qualcomm® QCM6690	31.862 ms	4 - 210 MB	NPU
YOLO26-Pose	ONNX	w8a16	Qualcomm® QCS7790	10.248 ms	4 - 213 MB	NPU
YOLO26-Pose	ONNX	w8a16	Qualcomm® QCS8750	2.821 ms	0 - 93 MB	NPU
YOLO26-Pose	ONNX	w8a16	Qualcomm® QCS7181	6.658 ms	179 - 179 MB	NPU
YOLO26-Pose	QNN_DLC	float	Snapdragon® X2 Elite	3.364 ms	5 - 5 MB	NPU
YOLO26-Pose	QNN_DLC	float	Snapdragon® X Elite	6.398 ms	5 - 5 MB	NPU
YOLO26-Pose	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	4.127 ms	5 - 197 MB	NPU
YOLO26-Pose	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	10.262 ms	4 - 214 MB	NPU
YOLO26-Pose	QNN_DLC	float	Qualcomm® QCS8275	16.752 ms	1 - 167 MB	NPU
YOLO26-Pose	QNN_DLC	float	Qualcomm® QCS8550 (Proxy)	5.764 ms	5 - 6 MB	NPU
YOLO26-Pose	QNN_DLC	float	Qualcomm® QCS8450	10.262 ms	4 - 214 MB	NPU
YOLO26-Pose	QNN_DLC	float	Snapdragon® 8 Elite Mobile	3.081 ms	0 - 163 MB	NPU
YOLO26-Pose	QNN_DLC	float	Qualcomm® SA8295P	10.509 ms	0 - 185 MB	NPU
YOLO26-Pose	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 Mobile	2.595 ms	5 - 193 MB	NPU
YOLO26-Pose	QNN_DLC	float	Qualcomm® SA7255P	16.752 ms	1 - 167 MB	NPU
YOLO26-Pose	QNN_DLC	float	Qualcomm® QCS9075	7.707 ms	5 - 13 MB	NPU
YOLO26-Pose	QNN_DLC	float	Qualcomm® QCS8750	3.081 ms	0 - 163 MB	NPU
YOLO26-Pose	QNN_DLC	float	Qualcomm® QCS7181	6.398 ms	5 - 5 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Snapdragon® X2 Elite	2.74 ms	2 - 2 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Snapdragon® X Elite	5.429 ms	2 - 2 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Snapdragon® 8 Gen 3 Mobile	3.226 ms	2 - 215 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Qualcomm® QCS8275	9.155 ms	1 - 186 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Qualcomm® QCS8550 (Proxy)	4.952 ms	0 - 17 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Qualcomm® QCS9075	6.049 ms	0 - 5 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	2.328 ms	2 - 199 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Snapdragon® 8 Elite Mobile	2.592 ms	2 - 189 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Snapdragon® 7 Gen 4 Mobile	6.061 ms	2 - 196 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Qualcomm® QCM6690	23.849 ms	2 - 189 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Qualcomm® SA7255P	9.155 ms	1 - 186 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Qualcomm® QCS7790	6.061 ms	2 - 196 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Qualcomm® QCS8750	2.592 ms	2 - 189 MB	NPU
YOLO26-Pose	QNN_DLC	w8a16	Qualcomm® QCS7181	5.429 ms	2 - 2 MB	NPU
YOLO26-Pose	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	5.36 ms	2 - 225 MB	NPU
YOLO26-Pose	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	14.425 ms	2 - 234 MB	NPU
YOLO26-Pose	TFLITE	float	Qualcomm® QCS8275	18.524 ms	2 - 190 MB	NPU
YOLO26-Pose	TFLITE	float	Qualcomm® QCS8550 (Proxy)	7.715 ms	0 - 3 MB	NPU
YOLO26-Pose	TFLITE	float	Qualcomm® SA8775P	24.215 ms	0 - 32 MB	GPU
YOLO26-Pose	TFLITE	float	Qualcomm® SA8650P	24.215 ms	0 - 32 MB	GPU
YOLO26-Pose	TFLITE	float	Qualcomm® SA8255P	24.215 ms	0 - 32 MB	GPU
YOLO26-Pose	TFLITE	float	Qualcomm® QCS8450	14.425 ms	2 - 234 MB	NPU
YOLO26-Pose	TFLITE	float	Snapdragon® 8 Elite Mobile	3.842 ms	0 - 181 MB	NPU
YOLO26-Pose	TFLITE	float	Qualcomm® SA8295P	12.614 ms	0 - 204 MB	NPU
YOLO26-Pose	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	3.229 ms	0 - 189 MB	NPU
YOLO26-Pose	TFLITE	float	Qualcomm® SA7255P	18.524 ms	2 - 190 MB	NPU
YOLO26-Pose	TFLITE	float	Qualcomm® QCS9075	9.303 ms	1 - 18 MB	NPU
YOLO26-Pose	TFLITE	float	Qualcomm® QCS8750	3.842 ms	0 - 181 MB	NPU

License

• The license for the original implementation of YOLO26-Pose can be found here.

References

• Ultralytics YOLO26 Docs: Pose Estimation
• 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.