Skip to content
Evaluation sources

Design Arena · visual design

Design Arena / Intelligence · Humans blindly select design work for web pages, interfaces, charts and slides.

Official evaluation
In NewsObserving
Evidence budgetNot scored
Upstream data as ofTo be confirmed
Last syncedNot collected yet

What it measures, and how

Compares only the base models of Models Arena; each design task shares a family budget, sub-tags are not treated as multiple independent sources, and votes are not double-counted with the codecategories total.

How this evidence is used

Under observation: the formal API protocol and display terms are clear; awaiting approved access and item-by-item verification of raw scores. Not yet collected, not yet scored.

Limits and data attribution

A public web page is no substitute for API authorisation; models, agents and builders cannot be mixed.

Data licence: Official API data permits personal and commercial display, with attribution and a visible link to the official site; API access requires an approved application.

Scores published by Design Arena / Intelligence. Raw scores and the News consensus score use different scales and cannot be added directly.