OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Yolox

Yolox is the repository for YOLOX, an anchor-free YOLO object-detection model with a PyTorch implementation. It includes code and documented workflows for image and video inference, training, and deployment through named inference ecosystems.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Yolox — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want an editable object-detection codebase and documented examples for inference or training, rather than a hosted detection interface.

Who it is for

  • Computer-vision developers integrating object detection into an application or pipeline
  • Researchers and practitioners training or comparing object-detection models
  • Engineers evaluating model deployment through ONNX, TensorRT, ncnn or OpenVINO

Documented uses

  • Run object detection on an image or video using a pretrained checkpoint.
  • Train YOLOX model variants using the documented COCO workflow.
  • Compare model variants using the repository’s reported benchmark measures.
  • Inspect assignment visualization during training or debugging.
  • Build an inference workflow around a listed deployment ecosystem.
Before you choose

Requirements & limits

  • The primary implementation in this repository is PyTorch; a separate MegEngine implementation is also referenced.
  • The README lists YOLOX-s, -m, -l and -x as standard models, and Nano and Tiny as light models, with benchmark measures to compare.
  • The demo workflow requires a checkpoint, detector configuration and input media; it exposes CPU/GPU selection and detection parameters.
  • The README documents COCO training and multi-machine training, while caching images requires substantial system RAM.
  • Apache-2.0 is the repository’s stated license.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Run the pretrained image demo

Install YOLOX from source and use its documented demo command to process an image.

Before you start

  • Python 3 and pip3
  • The YOLOX repository
  • A pretrained checkpoint from the repository’s benchmark table
  • An image file; the example uses assets/dog.jpg

Steps

  1. Clone the YOLOX repository and change into its directory.

  2. Install the project in editable mode with pip3 install -v -e . or use python3 setup.py develop.

  3. Download a pretrained model checkpoint and select a detector configuration, such as yolox-s.

  4. Run the documented image demo with the checkpoint, image path and configuration; choose CPU or GPU and set the provided detection options as needed.

Expected outcome

The demo processes the selected image and can save a result when --save_result is used.

Guide limitations

  • The repository’s example command requires the user to supply the checkpoint and input path.
  • The listed deployment runtimes do not by themselves establish compatibility with a particular hardware target or configuration.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
15 October 2026
Original website title
Megvii-BaseDetection/YOLOX · Official homepage title · retained in its source language
How we review sourcesSuggest a correction