Skip to content
AI frontend advisor
This project on GitHub, opens in a new tab

Find the front-end library that fits your project

The answers come from eight UI libraries, each built and measured on the same screen.

Enter sends, Shift+Enter adds a line

Or start with one of these

Rather read the numbers yourself? See the results for all eight builds.

What the answers come from

The advisor reads these files and nothing else. It names no overall winner, and when you describe your project it can order a short list of two or three against the needs you state.
  • The write-up

    write-up/README.md

    The picks by situation and the reasons behind them.

  • The screen spec

    spec/screen-spec.md

    The one screen every library had to build.

  • The results

    results/*.json

    One measured result file for each of the eight builds.

How it works

  1. Tell it about your project

    The framework, the screens you are building, and what matters most: bundle size, accessibility, or speed of work.

  2. Get a short list

    Two or three libraries in order against the needs you stated, with the need that decided the order.

  3. Check the evidence

    Every answer links to the results, the write-up, or the screen spec, so you can read the source yourself.

What you are looking at

  • The advisor drawer asking about framework, bundle weight, accessibility and how finished the components should be

    The advisor asks about your project before it names anything.

  • The tickets screen built with shadcn/ui, at 1440 pixels wide, with its filters, table and pagination

    The screen every library built, here in shadcn/ui. All of them are live on this site.

A screenshot opens at full size. The tickets on every screen are made up: 240 rows from a seeded generator, the same rows for every library. Nothing you type into a screen is saved, and the next visitor sees the same data you do.

The eight libraries

React

Vue

Read the evidence

Ready to pick one?

Describe your project at the top of this page, or read how the comparison was run.

Ask the advisorHow this works