# DEIM-S

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

- Family: DEIM (Intellindust AI Lab)
- Parameters: 10.0M
- GFLOPs: 25.0
- Default input size: 640px
- Detection: detr · nms-free
- Architecture: transformer · HGNetv2 backbone
- Weight license: Apache-2.0
- Paper-reported mAP@50-95: 49%
- Paper: https://arxiv.org/abs/2412.04234
- Original code: https://github.com/Intellindust-AI-Lab/DEIM

## Benchmarks (COCO val2017)

| Hardware | Runtime | mAP@50-95 | FPS | Latency (ms) | VRAM (MB) |
|---|---|---|---|---|---|
| NVIDIA Jetson Orin Nano Super 8GB | ONNX Runtime FP32 | 52.1 | 10.4 | 96.2 | - |
| NVIDIA Jetson Orin Nano Super 8GB | PyTorch FP32 | 52.1 | 10.2 | 98.1 | 139 |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP16 | 52.0 | 29.9 | 33.4 | - |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP32 | 52.1 | 22.2 | 45.0 | - |
| NVIDIA RTX 5070 Ti | ONNX Runtime FP32 | 52.1 | 67.5 | 14.8 | - |
| NVIDIA RTX 5070 Ti | PyTorch FP32 | 52.1 | 33.4 | 30.0 | 142 |
| NVIDIA RTX 5070 Ti | TensorRT FP16 | 51.9 | 82.4 | 12.1 | - |
| NVIDIA RTX 5070 Ti | TensorRT FP32 | 52.1 | 97.6 | 10.2 | - |
| Raspberry Pi 5 | ONNX Runtime FP32 | 52.1 | 1.8 | 559.2 | - |
| Raspberry Pi 5 | PyTorch FP32 | 52.1 | 1.1 | 878.2 | - |

## Usage with LibreYOLO

```python
from libreyolo import LibreYOLO

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

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