YOLO26-Segmentation: Optimized for Qualcomm Devices

Ultralytics YOLO26 is a machine learning model that predicts bounding boxes, segmentation masks and classes of objects in an image.

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

Model Details

Model Type: Model_use_case.semantic_segmentation

Model Stats:

• Model checkpoint: YOLO26N-Seg
• Input resolution: 640x640
• Number of output classes: 80
• Number of parameters: 2.7M
• Model size (float): 9.29 MB
• Model size (w8a16): 5.25 MB

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
YOLO26-Segmentation	ONNX	float	Snapdragon® 8 Elite Gen 5 Mobile	4.074 ms	15 - 189 MB	NPU
YOLO26-Segmentation	ONNX	float	Snapdragon® X2 Elite	3.916 ms	15 - 15 MB	NPU
YOLO26-Segmentation	ONNX	float	Snapdragon® X Elite	7.312 ms	17 - 17 MB	NPU
YOLO26-Segmentation	ONNX	float	Snapdragon® 8 Gen 3 Mobile	4.75 ms	16 - 228 MB	NPU
YOLO26-Segmentation	ONNX	float	Qualcomm® QCS8550 (Proxy)	6.886 ms	11 - 20 MB	NPU
YOLO26-Segmentation	ONNX	float	Qualcomm® QCS9075	9.957 ms	12 - 15 MB	NPU
YOLO26-Segmentation	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	4.552 ms	11 - 185 MB	NPU
YOLO26-Segmentation	ONNX	w8a16	Snapdragon® 8 Elite Gen 5 Mobile	2.642 ms	0 - 94 MB	NPU
YOLO26-Segmentation	ONNX	w8a16	Snapdragon® X2 Elite	2.874 ms	6 - 6 MB	NPU
YOLO26-Segmentation	ONNX	w8a16	Snapdragon® X Elite	6.903 ms	7 - 7 MB	NPU
YOLO26-Segmentation	ONNX	w8a16	Snapdragon® 8 Gen 3 Mobile	3.883 ms	9 - 247 MB	NPU
YOLO26-Segmentation	ONNX	w8a16	Qualcomm® QCS6490	444.357 ms	171 - 176 MB	CPU
YOLO26-Segmentation	ONNX	w8a16	Qualcomm® QCS8550 (Proxy)	6.282 ms	8 - 13 MB	NPU
YOLO26-Segmentation	ONNX	w8a16	Qualcomm® QCS9075	7.143 ms	8 - 11 MB	NPU
YOLO26-Segmentation	ONNX	w8a16	Qualcomm® QCM6690	215.887 ms	170 - 181 MB	CPU
YOLO26-Segmentation	ONNX	w8a16	Snapdragon® 8 Elite For Galaxy Mobile	2.984 ms	0 - 93 MB	NPU
YOLO26-Segmentation	ONNX	w8a16	Snapdragon® 7 Gen 4 Mobile	200.226 ms	105 - 115 MB	CPU
YOLO26-Segmentation	TFLITE	float	Snapdragon® 8 Elite Gen 5 Mobile	2.603 ms	0 - 169 MB	NPU
YOLO26-Segmentation	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	3.859 ms	3 - 196 MB	NPU
YOLO26-Segmentation	TFLITE	float	Qualcomm® QCS8275 (Proxy)	16.925 ms	4 - 167 MB	NPU
YOLO26-Segmentation	TFLITE	float	Qualcomm® QCS8550 (Proxy)	5.174 ms	4 - 7 MB	NPU
YOLO26-Segmentation	TFLITE	float	Qualcomm® SA8775P	6.776 ms	4 - 170 MB	NPU
YOLO26-Segmentation	TFLITE	float	Qualcomm® QCS9075	7.871 ms	4 - 23 MB	NPU
YOLO26-Segmentation	TFLITE	float	Qualcomm® QCS8450 (Proxy)	12.002 ms	4 - 213 MB	NPU
YOLO26-Segmentation	TFLITE	float	Qualcomm® SA7255P	16.925 ms	4 - 167 MB	NPU
YOLO26-Segmentation	TFLITE	float	Qualcomm® SA8295P	11.058 ms	5 - 184 MB	NPU
YOLO26-Segmentation	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	3.224 ms	0 - 171 MB	NPU

License

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

References

• Ultralytics YOLO26 Docs: Instance Segmentation
• 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.