Parameters31.0M
GFLOPs101.0
Input Size640px
Best mAP60.1%
LicenseApache-2.0
Architecture
Type
transformer
Backbone
ECViT (compact ViT)
Neck
HybridEncoder
Head
DETR
Benchmark Results
Performance on COCO val2017 across different hardware configurations
| Hardware | Runtime | mAP@50-95 | FPS | Latency | VRAM |
|---|---|---|---|---|---|
| NVIDIA Jetson Orin Nano Super 8GB | ONNX Runtime FP32 | 60.1% | 2.7 | 367.3ms | - |
| NVIDIA Jetson Orin Nano Super 8GB | PyTorch FP32 | 60.1% | 3.4 | 296.0ms | 212 MB |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP16 | 60.0% | 6.2 | 161.5ms | - |
| NVIDIA Jetson Orin Nano Super 8GB | TensorRT FP32 | 60.1% | 6.1 | 163.8ms | - |
| NVIDIA RTX 5070 Ti | ONNX Runtime FP32 | 60.1% | 33.6 | 29.8ms | - |
| NVIDIA RTX 5070 Ti | PyTorch FP32 | 60.1% | 24.2 | 41.3ms | 215 MB |
| NVIDIA RTX 5070 Ti | TensorRT FP16 | 60.0% | 41.7 | 24.0ms | - |
| NVIDIA RTX 5070 Ti | TensorRT FP32 | 60.1% | 49.0 | 20.4ms | - |
Speed Breakdown(NVIDIA Jetson Orin Nano Super 8GB)
16.4ms
276.5ms
3.1ms
Preprocess
Inference
Postprocess (NMS)
Usage with LibreYOLO
from libreyolo import LibreYOLO
# Load model (auto-downloads from HuggingFace if not found locally)
model = LibreYOLO("LibreECl.pt")
# Run inference
result = model("image.jpg", conf=0.25, iou=0.45)
# Process results
print(f"Found {len(result)} objects")
print(result.boxes.xyxy) # bounding boxes (N, 4)
print(result.boxes.conf) # confidence scores (N,)
print(result.boxes.cls) # class IDs (N,)detrnms-freePaper: 57% mAP