Back to Leaderboard
YOLOv11-M
yolov11one-stage detector with C3K2 backbone
Parameters
20.1M
FLOPs
68.0G
Input Size
640px
License
MIT
Architecture
Type
one-stage
Backbone
C3K2
Neck
SPPF
Head
Decoupled
Benchmark Results
Performance on COCO val2017 across different hardware configurations
| Hardware | mAP@50-95 | FPS | Latency | VRAM |
|---|---|---|---|---|
| NVIDIA A100 (TensorRT FP16) | 51.6% | 115.0 | 8.7ms | 939 MB |
| NVIDIA T4 (TensorRT FP16) | 51.6% | 49.1 | 20.4ms | 967 MB |
| CPU (ONNX Runtime) | 51.4% | 5.5 | 180.3ms | 980 MB |
Speed Breakdown (A100 TensorRT)
End-to-end latency breakdown showing preprocessing, inference, and postprocessing times
1.4ms
4.7ms
2.6ms
Preprocess
Inference
Postprocess (NMS)
Usage with LibreYOLO
from libreyolo import YOLO
# Load model
model = YOLO.from_pretrained("https://huggingface.co/Libre-YOLO/yolov11m")
# Run inference
results = model.predict("image.jpg")
# Process results
for box in results.boxes:
print(f"Class: {box.cls}, Confidence: {box.conf:.2f}")latestbalanced