# RT-DETR-R101

> RT-DETR-R101: transformer object detector from the RT-DETR family. Available in LibreYOLO, the MIT-licensed open-source library (free for commercial use).

- Family: RT-DETR (Baidu)
- Parameters: 76.0M
- GFLOPs: 259.0
- Default input size: 640px
- Detection: detr · nms-free
- Architecture: transformer · ResNet-101 backbone
- Weight license: Apache-2.0
- Paper-reported mAP@50-95: 54.3%
- Paper: https://arxiv.org/abs/2304.08069
- Original code: https://github.com/lyuwenyu/RT-DETR

## Benchmarks (COCO val2017)

| Hardware | Runtime | mAP@50-95 | FPS | Latency (ms) | VRAM (MB) |
|---|---|---|---|---|---|
| NVIDIA A100 | PyTorch FP32 | 53.9 | 18.6 | 53.6 | 508 |
| NVIDIA Jetson Orin Nano Super 8GB | ONNX Runtime FP32 | 56.8 | 2.9 | 343.9 | - |
| NVIDIA Jetson Orin Nano Super 8GB | PyTorch FP32 | 56.8 | 2.8 | 359.6 | 450 |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP16 | 56.8 | 17.0 | 58.8 | - |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP32 | 56.8 | 8.4 | 118.5 | - |
| NVIDIA RTX 5070 Ti | ONNX Runtime FP32 | 56.8 | 47.5 | 21.1 | - |
| NVIDIA RTX 5070 Ti | PyTorch FP32 | 56.8 | 23.6 | 42.4 | 456 |
| NVIDIA RTX 5070 Ti | TensorRT FP16 | 56.7 | 88.0 | 11.4 | - |
| NVIDIA RTX 5070 Ti | TensorRT FP32 | 56.8 | 64.4 | 15.5 | - |

## Usage with LibreYOLO

```python
from libreyolo import LibreYOLO

model = LibreYOLO("LibreRTDETRr101.pt")
result = model("image.jpg", conf=0.25, iou=0.45)
```

Source: https://www.visionanalysis.org/model/rtdetr-r101. Benchmarks produced with LibreYOLO (https://github.com/Libre-YOLO/libreyolo).