Back to Leaderboard
Parameters
54.2M
FLOPs
155.6G
Input Size
640px
License
Apache-2.0
Architecture
Type
one-stage
Backbone
CSPDarknet
Neck
PAFPN
Head
Decoupled
Benchmark Results
Performance on COCO val2017 across different hardware configurations
| Hardware | mAP@50-95 | FPS | Latency | VRAM |
|---|---|---|---|---|
| NVIDIA A100 (TensorRT FP16) | 49.6% | 77.0 | 13.0ms | 2375 MB |
| NVIDIA T4 (TensorRT FP16) | 49.8% | 29.0 | 34.5ms | 2391 MB |
| CPU (ONNX Runtime) | 49.6% | 3.4 | 298.1ms | 2229 MB |
Speed Breakdown (A100 TensorRT)
End-to-end latency breakdown showing preprocessing, inference, and postprocessing times
1.1ms
9.3ms
2.6ms
Preprocess
Inference
Postprocess (NMS)
Usage with LibreYOLO
from libreyolo import YOLO
# Load model
model = YOLO.from_pretrained("https://huggingface.co/Libre-YOLO/yolox-l")
# Run inference
results = model.predict("image.jpg")
# Process results
for box in results.boxes:
print(f"Class: {box.cls}, Confidence: {box.conf:.2f}")high-accuracymegvii
Related Models (yolox)