# YOLO11-N

> YOLO11-N: one-stage object detector from the YOLO11 family. Planned for LibreYOLO (not yet packaged).

- Family: YOLO11 (External baseline)
- Parameters: 2.6M
- GFLOPs: 6.5
- Default input size: 640px
- Detection: anchor-free · nms
- Architecture: one-stage · CSPDarknet backbone
- Weight license: non-permissive
- Paper-reported mAP@50-95: 39.5%

## Benchmarks (COCO val2017)

| Hardware | Runtime | mAP@50-95 | FPS | Latency (ms) | VRAM (MB) |
|---|---|---|---|---|---|
| NVIDIA Jetson Orin Nano Super 8GB | ONNX Runtime FP32 | 38.7 | 8.2 | 121.5 | - |
| NVIDIA Jetson Orin Nano Super 8GB | PyTorch FP32 | 38.7 | 22.9 | 43.8 | 66 |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP16 | 38.7 | 52.6 | 19.0 | 14 |
| Raspberry Pi 5 | ncnn FP32 | 38.7 | 12.2 | 81.7 | - |
| Raspberry Pi 5 | ONNX Runtime FP32 | 38.7 | 6.2 | 162.2 | - |
| Raspberry Pi 5 | PyTorch FP32 | 38.7 | 4.9 | 204.7 | - |


Source: https://www.visionanalysis.org/model/yolo11n. Benchmarks produced with LibreYOLO (https://github.com/Libre-YOLO/libreyolo).