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

Edgeyolo

EdgeYOLO is an open-source, anchor-free object-detection model implemented in Python for real-time computer vision on edge devices. Its public repository provides model variants and pretrained weights, plus resources for inference, training, evaluation and deployment across selected runtimes and hardware platforms.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Edgeyolo — 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 EdgeYOLO when you need an object detector implementation to run or adapt in an edge-oriented computer-vision pipeline, and can manage the local environment and any platform-specific dependencies.

Who it is for

  • Computer-vision developers building embedded or edge-device applications.
  • Teams evaluating object detection on COCO2017 or aerial imagery using VisDrone2019-DET.
  • Developers adapting model exports and deployment examples for supported platform-specific runtimes.

Documented uses

  • Detect objects in image and video inputs.
  • Evaluate object-detection models on COCO2017 or VisDrone2019-DET.
  • Train or export models and investigate deployment using documented runtime and hardware paths.
Before you choose

Requirements & limits

The implementation is publicly available on GitHub under the Apache-2.0 license. It is self-hosted rather than a hosted inference service.

  • The repository provides multiple model sizes, with different reported accuracy and speed figures.
  • The project reports benchmarks on NVIDIA Jetson AGX Xavier; performance depends on the documented test settings and should not be treated as a guarantee for other hardware.
  • Platform deployment paths have distinct dependencies; TensorRT requires torch2trt and the TensorRT Development Toolkit, while the Ascend conversion path requires vendor libraries and tools.
  • The README reports RKNN inference deployment code specifically for RK3588.

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

Install the repository's Python requirements

Prepare a local checkout for the documented inference and development workflow.

Before you start

  • Git
  • Python environment capable of installing the repository requirements

Steps

  1. Clone https://github.com/LSH9832/edgeyolo.git.

  2. Change into the edgeyolo repository directory.

  3. Install dependencies with pip install -r requirements.txt.

Expected outcome

The repository dependencies are installed for the next project workflow.

Guide limitations

  • This setup sequence does not by itself download model weights or run inference.
  • TensorRT use has additional requirements: torch2trt and TensorRT Development Toolkit version greater than 7.1.3.0.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
LSH9832/edgeyolo · Official homepage title · retained in its source language
How we review sourcesSuggest a correction