# YOLOv9-C

> YOLOv9-C: one-stage object detector from the YOLOv9 family. Available in LibreYOLO, the MIT-licensed open-source library (free for commercial use).

- Family: YOLOv9 (Academia Sinica)
- Parameters: 25.5M
- GFLOPs: 51.8
- Default input size: 640px
- Detection: anchor-free · nms
- Architecture: one-stage · GELAN backbone
- Weight license: MIT
- Paper-reported mAP@50-95: 53%
- Paper: https://arxiv.org/abs/2402.13616
- Original code: https://github.com/MultimediaTechLab/YOLO

## Benchmarks (COCO val2017)

| Hardware | Runtime | mAP@50-95 | FPS | Latency (ms) | VRAM (MB) |
|---|---|---|---|---|---|
| NVIDIA Jetson Orin Nano Super 8GB | ONNX Runtime FP32 | 57.1 | 6.4 | 156.1 | - |
| NVIDIA Jetson Orin Nano Super 8GB | PyTorch FP32 | 57.1 | 6.4 | 155.6 | 239 |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP16 | 57.1 | 24.9 | 40.1 | - |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP32 | 57.1 | 13.2 | 76.0 | - |
| NVIDIA RTX 5070 Ti | ONNX Runtime FP32 | 57.1 | 63.5 | 15.7 | - |
| NVIDIA RTX 5070 Ti | PyTorch FP32 | 57.1 | 42.3 | 23.6 | 241 |
| NVIDIA RTX 5070 Ti | TensorRT FP16 | 57.2 | 74.7 | 13.4 | - |
| NVIDIA RTX 5070 Ti | TensorRT FP32 | 57.1 | 76.7 | 13.0 | - |

## Usage with LibreYOLO

```python
from libreyolo import LibreYOLO

model = LibreYOLO("LibreYOLO9c.pt")
result = model("image.jpg", conf=0.25, iou=0.45)
```

Source: https://www.visionanalysis.org/model/yolov9c. Benchmarks produced with LibreYOLO (https://github.com/Libre-YOLO/libreyolo).