The AI Stack
Sign in

CVAT

Annotates images, video, and 3D data for computer vision teams

Updated June 2026

Overview

Website
cvat.ai
Founded
2017
Headquarters
Wilmington, Delaware, USA
Segment
Data Labeling & Annotation

Product overview

CVAT is an open-source visual data annotation platform used to label images, video, and 3D data for computer vision and AI workflows. It includes automation tools for detection, segmentation, and tracking, plus collaboration and QA features for teams.

Revenue model

Open-source platform with enterprise, services, consulting, and customization offerings

Moat

  • Switching Costs
  • Distribution
  • Network Effects

CVAT’s moat is mainly open-source adoption and workflow integration: it becomes sticky when teams build annotation processes, scripts, and QA around it. It also benefits from network/distribution effects via community contributions and enterprise familiarity, but it is not strongly protected by patents or proprietary technology.

Headwinds

Open-source model limits monetization opportunities compared to commercial competitors