HeroUI v3 Lands as a Ground-Up Rewrite for React and React Native, Built on Tailwind CSS v4

· Source: InfoQ · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning · Depth: Intermediate, quick

Summary

HeroUI v3, a React component library formerly known as NextUI, has been released as a ground-up rewrite for web and native applications. Launched in March 2026, it features over 75 web components, including 21 new ones, and a new React Native library with 37 components. The library is built on React Aria Components for accessibility and Tailwind CSS v4 for styling, utilizing CSS variables, OKLCH colours, and BEM modifiers. All animations are now CSS-driven without JavaScript runtime, and it adopts a compound component API compatible with React 19 and Next.js. HeroUI v3 also integrates AI-assisted development via an MCP Server, Agent Skills, and an llms.txt file for coding assistants, alongside a Figma Kit v3. The core "@heroui/react" package sees around 341,000 weekly npm downloads, with "heroui-native" adding 11,000, and its GitHub repository has nearly 27,700 stars. Recent point releases, including v3.2.0 on June 16, 2026, maintain a monthly minor release cadence.

Key takeaway

For AI Architects or UI Development Managers evaluating component libraries, HeroUI v3 provides a modern, unified solution for React and React Native. Built on Tailwind CSS v4 and React Aria, it offers strong accessibility and styling. If migrating from v2, plan carefully; v2 and v3 are incompatible, and some components are removed. Utilize the official migration guide. Its AI-assisted development features can streamline your team's workflow, but assess if missing v2 features affect your project.

Key insights

HeroUI v3 is a comprehensive rewrite, integrating modern web tech and AI tools for a unified React and React Native design system.

Principles

Method

HeroUI v3's rewrite involved building on React Aria Components and Tailwind CSS v4, moving animations to CSS, adopting a compound component API, and integrating AI-assisted development tools like MCP Server and llms.txt.

In practice

Topics

Code references

Best for: Software Engineer, AI Engineer, AI Architect

Related on AIssential

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by InfoQ.