Arena WebDev

In plain words · Which AI built the websites and apps people preferred

What does it measure?

A ranking of which AI builds the more usable result when asked to make a website or app. People pick by looking at the running page rather than grading code, so a high score means the AI is good at producing something that looks right and works.

Who evaluates it, and how?

In the WebDev section of Arena Intelligence, a user describes a page or app, two anonymous AIs write the code, and the results are shown running side by side. The user votes for the better one, and votes are scored with the same statistical model as text Arena.

Reading the score

This is a head-to-head record, not an exam score. It is computed from the probability of being preferred over other AIs, so there is no maximum; a 100-point gap means the leader is preferred about 64% of the time. Gaps of a few dozen points are effectively the same tier even if the rank differs. Values in this table are currently about 1,200–1,800. First impressions of the running page drive the vote, so code maintainability and security are not reflected.

Last updated: 2026-10-01