# RF-DETR-N

> RF-DETR-N: transformer object detector from the RF-DETR family. Available in LibreYOLO, the MIT-licensed open-source library (free for commercial use).

- Family: RF-DETR (Roboflow)
- Parameters: 30.5M
- GFLOPs: 0.0
- Default input size: 384px
- Detection: detr · nms-free
- Architecture: transformer · DINOv2 backbone
- Weight license: Apache-2.0
- Paper-reported mAP@50-95: 48.4%
- Paper: https://blog.roboflow.com/rf-detr
- Original code: https://github.com/roboflow/rf-detr

## Benchmarks (COCO val2017)

| Hardware | Runtime | mAP@50-95 | FPS | Latency (ms) | VRAM (MB) |
|---|---|---|---|---|---|
| NVIDIA Jetson Orin Nano Super 8GB | ONNX Runtime FP32 | 51.4 | 10.4 | 96.0 | - |
| NVIDIA Jetson Orin Nano Super 8GB | PyTorch FP32 | 51.4 | 10.2 | 98.3 | 150 |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP16 | 51.3 | 25.4 | 39.4 | - |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP32 | 51.4 | 23.7 | 42.1 | - |
| NVIDIA RTX 5070 Ti | ONNX Runtime FP32 | 51.4 | 65.7 | 15.2 | - |
| NVIDIA RTX 5070 Ti | PyTorch FP32 | 51.4 | 39.5 | 25.3 | 151 |
| NVIDIA RTX 5070 Ti | TensorRT FP16 | 51.4 | 58.3 | 17.1 | - |
| NVIDIA RTX 5070 Ti | TensorRT FP32 | 51.4 | 67.6 | 14.8 | - |
| Raspberry Pi 5 | ONNX Runtime FP32 | 51.4 | 2.4 | 409.2 | - |
| Raspberry Pi 5 | PyTorch FP32 | 51.4 | 2.1 | 479.6 | - |

## Usage with LibreYOLO

```python
from libreyolo import LibreYOLO

model = LibreYOLO("LibreRFDETRn.pt")
result = model("image.jpg", conf=0.25, iou=0.45)
```

Source: https://www.visionanalysis.org/model/rfdetr-n. Benchmarks produced with LibreYOLO (https://github.com/Libre-YOLO/libreyolo).