YOLOX: Anchor-free object detection model
PyTorch anchor-free object detector with scalable model variants.
Learn more about YOLOX
YOLOX is an object detection model that removes the anchor-based detection mechanism used in earlier YOLO versions, replacing it with an anchor-free approach. The architecture is implemented in PyTorch with a companion MegEngine implementation available separately. The model family includes variants ranging from YOLOX-Nano with 0.91M parameters to YOLOX-X with 99.1M parameters, enabling deployment across different computational constraints. Common applications include real-time object detection in computer vision pipelines, with support for edge deployment through quantization and conversion to optimized inference formats.
Anchor-Free Detection
Eliminates anchor boxes from the detection pipeline, removing hyperparameter tuning for anchor dimensions and aspect ratios. Simplifies training workflow while maintaining accuracy comparable to anchor-based YOLO versions.
Multi-Framework Deployment
Export trained models to ONNX, TensorRT, ncnn, OpenVINO, and MegEngine with built-in conversion scripts. Enables deployment across cloud GPUs, edge devices, and mobile hardware without manual optimization.
Scalable Model Variants
Provides five model sizes from Nano (0.91M parameters) to X (99.1M parameters) with consistent architecture. Allows selection based on target hardware constraints, from resource-limited embedded devices to high-performance server inference.
import torch
from yolox.exp import get_exp
exp = get_exp(None, "yolox-s")
model = exp.get_model()
ckpt = torch.load("yolox_s.pth", map_location="cpu")
model.load_state_dict(ckpt["model"])
model.eval()
img = torch.randn(1, 3, 640, 640)
outputs = model(img)YOLOX 0.3.0 adds torch.hub integration, just-in-time compilation, wandb logging support, and various demo and deployment improvements.
- –support loading YOLOX model through torch.hub
- –support just-in-time compile op
- –support wandb logger
- –support freeze function for torch module
- –support showing YOLOX demo in a live window
This release makes YOLOX installable via pip and includes significant memory optimizations and evaluation improvements.
- –Saved 30% memory useage in COCO training
- –Log per class AP & AP during evaluation
- –Users could install yolox from pip now! Supports on more platform is coming
- –Optimize dynamic matching in label assignment
This release introduces 2x faster training through optimized preprocessing and image caching, removes Apex dependency in favor of torch amp.
- –Support image caching for faster training, which requires large system RAM
- –Remove the dependence of apex and support torch amp training
- –Optimize the preprocessing for faster training
- –Replace the older distort augmentation with new HSV aug for faster training and better performance
See how people are using YOLOX
Related Repositories
Discover similar tools and frameworks used by developers
OpenClaw
Personal AI assistant that runs on your own devices and connects to messaging platforms like WhatsApp, Telegram, and Slack.
DeepFace
Python library wrapping multiple face recognition deep learning models.
whisper.cpp
Lightweight C++ port of OpenAI Whisper for cross-platform speech recognition.
Triton
Domain-specific language and compiler for writing GPU deep learning primitives with higher productivity than CUDA.
ByteTrack
Multi-object tracker associating low-confidence detections across frames.