Back to Leaderboard

YOLOX-Tiny

yolox

one-stage detector with CSPDarknet backbone

Parameters

5.1M

FLOPs

6.5G

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
HardwaremAP@50-95FPSLatencyVRAM
NVIDIA A100 (TensorRT FP16)32.9%206.64.8ms276 MB
NVIDIA T4 (TensorRT FP16)32.8%86.211.6ms195 MB
CPU (ONNX Runtime)32.7%10.198.6ms242 MB
Speed Breakdown (A100 TensorRT)
End-to-end latency breakdown showing preprocessing, inference, and postprocessing times
1.3ms
1.3ms
2.3ms
Preprocess
Inference
Postprocess (NMS)
Usage with LibreYOLO
from libreyolo import YOLO

# Load model
model = YOLO.from_pretrained("https://huggingface.co/Libre-YOLO/yolox-tiny")

# Run inference
results = model.predict("image.jpg")

# Process results
for box in results.boxes:
    print(f"Class: {box.cls}, Confidence: {box.conf:.2f}")
lightweightedge