# ECDet-M

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

- Family: ECDet (Intellindust AI Lab)
- Parameters: 18.0M
- GFLOPs: 53.0
- Default input size: 640px
- Detection: detr · nms-free
- Architecture: transformer · ECViT (compact ViT) backbone
- Weight license: Apache-2.0
- Paper-reported mAP@50-95: 54.3%
- Paper: https://arxiv.org/abs/2603.18739
- Original code: https://github.com/Intellindust-AI-Lab/EdgeCrafter

## Benchmarks (COCO val2017)

| Hardware | Runtime | mAP@50-95 | FPS | Latency (ms) | VRAM (MB) |
|---|---|---|---|---|---|
| NVIDIA Jetson Orin Nano Super 8GB | ONNX Runtime FP32 | 58.4 | 3.9 | 254.4 | - |
| NVIDIA Jetson Orin Nano Super 8GB | PyTorch FP32 | 58.4 | 5.0 | 200.8 | 160 |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP16 | 58.3 | 8.3 | 119.7 | - |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP32 | 58.3 | 8.2 | 121.6 | - |
| NVIDIA RTX 5070 Ti | ONNX Runtime FP32 | 58.4 | 40.2 | 24.9 | - |
| NVIDIA RTX 5070 Ti | PyTorch FP32 | 58.4 | 26.5 | 37.7 | 162 |
| NVIDIA RTX 5070 Ti | TensorRT FP16 | 58.2 | 48.9 | 20.5 | - |
| NVIDIA RTX 5070 Ti | TensorRT FP32 | 58.4 | 59.2 | 16.9 | - |
| Raspberry Pi 5 | ONNX Runtime FP32 | 58.4 | 0.6 | 1708.2 | - |
| Raspberry Pi 5 | PyTorch FP32 | 58.4 | 0.5 | 1868.7 | - |

## Usage with LibreYOLO

```python
from libreyolo import LibreYOLO

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

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