Back to Leaderboard
Parameters
2.3M
FLOPs
6.7G
Input Size
640px
License
MIT
Architecture
Type
one-stage
Backbone
CSPDarknet
Neck
PAFPN
Head
NMS-Free
Benchmark Results
Performance on COCO val2017 across different hardware configurations
| Hardware | mAP@50-95 | FPS | Latency | VRAM |
|---|---|---|---|---|
| NVIDIA A100 (TensorRT FP16) | 38.4% | 380.4 | 2.6ms | 148 MB |
| NVIDIA T4 (TensorRT FP16) | 38.6% | 199.9 | 5.0ms | 252 MB |
| CPU (ONNX Runtime) | 38.5% | 28.1 | 35.6ms | 161 MB |
Speed Breakdown (A100 TensorRT)
End-to-end latency breakdown showing preprocessing, inference, and postprocessing times
1.3ms
1.3ms
0.1ms
Preprocess
Inference
Postprocess (NMS)
Usage with LibreYOLO
from libreyolo import YOLO
# Load model
model = YOLO.from_pretrained("https://huggingface.co/Libre-YOLO/yolov10n")
# Run inference
results = model.predict("image.jpg")
# Process results
for box in results.boxes:
print(f"Class: {box.cls}, Confidence: {box.conf:.2f}")nms-freereal-timeedge-friendly
Notes
No NMS required - fastest end-to-end