# YOLOX-Tiny

> YOLOX-Tiny: one-stage object detector from the YOLOX family. Available in LibreYOLO, the MIT-licensed open-source library (free for commercial use).

- Family: YOLOX (Megvii)
- Parameters: 5.1M
- GFLOPs: 7.7
- Default input size: 640px
- Detection: anchor-free · nms
- Architecture: one-stage · CSPDarknet backbone
- Weight license: Apache-2.0
- Paper-reported mAP@50-95: 32.8%
- Paper: https://arxiv.org/abs/2107.08430
- Original code: https://github.com/Megvii-BaseDetection/YOLOX

## Benchmarks (COCO val2017)

| Hardware | Runtime | mAP@50-95 | FPS | Latency (ms) | VRAM (MB) |
|---|---|---|---|---|---|
| NVIDIA Jetson Orin Nano Super 8GB | ONNX Runtime FP32 | 35.5 | 31.4 | 31.8 | - |
| NVIDIA Jetson Orin Nano Super 8GB | PyTorch FP32 | 35.5 | 20.3 | 49.2 | 37 |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP16 | 35.5 | 51.6 | 19.4 | - |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP32 | 35.5 | 43.6 | 23.0 | - |
| NVIDIA RTX 5070 Ti | ONNX Runtime FP32 | 35.5 | 76.2 | 13.1 | - |
| NVIDIA RTX 5070 Ti | PyTorch FP32 | 35.5 | 61.6 | 16.2 | 37 |
| NVIDIA RTX 5070 Ti | TensorRT FP16 | 35.4 | 97.7 | 10.2 | - |
| NVIDIA RTX 5070 Ti | TensorRT FP32 | 33.9 | 87.2 | 11.5 | - |

## Usage with LibreYOLO

```python
from libreyolo import LibreYOLO

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

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