RF-DETR: Optimized for Qualcomm Devices

DETR is a machine learning model that can detect objects (trained on COCO dataset).

This is based on the implementation of RF-DETR 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	float	Universal	QAIRT 2.50, ONNX Runtime 1.30.0	Download
QNN_DLC	float	Universal	QAIRT 2.50	Download

For more device-specific assets and performance metrics, visit RF-DETR 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 RF-DETR on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.object_detection

Model Stats:

• Input resolution: 512x512
• Model checkpoint: RF-DETR-small
• Model size (float): 109 MB
• Number of parameters: 28.5M
• Supported variants: nano (384x384), small (512x512), medium (576x576), base (560x560)

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
RF-DETR	ONNX	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	18.102 ms	11 - 303 MB	NPU
RF-DETR	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	21.279 ms	12 - 305 MB	NPU
RF-DETR	ONNX	float	Snapdragon® 8 Gen 3 Mobile	28.956 ms	0 - 391 MB	NPU
RF-DETR	ONNX	float	Snapdragon® 8 Gen 1 Mobile	68.282 ms	0 - 407 MB	NPU
RF-DETR	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	48.625 ms	9 - 16 MB	NPU
RF-DETR	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	38.513 ms	11 - 14 MB	NPU
RF-DETR	ONNX	float	Qualcomm® QCS8450	68.282 ms	0 - 407 MB	NPU
RF-DETR	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	47.751 ms	11 - 18 MB	NPU
RF-DETR	ONNX	float	Qualcomm® Dragonwing™ Q-8750	21.279 ms	12 - 305 MB	NPU
RF-DETR	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	18.553 ms	3 - 338 MB	NPU
RF-DETR	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	21.863 ms	3 - 334 MB	NPU
RF-DETR	QNN_DLC	float	Snapdragon® X2 Elite	19.233 ms	3 - 3 MB	NPU
RF-DETR	QNN_DLC	float	Snapdragon® X Elite	41.827 ms	3 - 3 MB	NPU
RF-DETR	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	29.022 ms	0 - 447 MB	NPU
RF-DETR	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	74.059 ms	0 - 449 MB	NPU
RF-DETR	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	47.163 ms	3 - 9 MB	NPU
RF-DETR	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	40.33 ms	3 - 5 MB	NPU
RF-DETR	QNN_DLC	float	Qualcomm® SA8775P	47.412 ms	1 - 330 MB	NPU
RF-DETR	QNN_DLC	float	Qualcomm® SA8650P	47.412 ms	1 - 330 MB	NPU
RF-DETR	QNN_DLC	float	Qualcomm® SA8255P	47.412 ms	1 - 330 MB	NPU
RF-DETR	QNN_DLC	float	Qualcomm® QCS8450	74.059 ms	0 - 449 MB	NPU
RF-DETR	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	47.776 ms	3 - 8 MB	NPU
RF-DETR	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	41.827 ms	3 - 3 MB	NPU
RF-DETR	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	21.863 ms	3 - 334 MB	NPU
RF-DETR	QNN_DLC	float	Qualcomm® SA8295P	69.1 ms	0 - 340 MB	NPU

License

• The license for the original implementation of RF-DETR can be found here.

References

• RF-DETR A SOTA Real-Time Object Detection Model
• 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.