How AA-3D works: build, check, vote, rank
Each model works as an agent in an offline sandbox and turns a written brief into a three.js web page, with up to three hours and 250 turns. The sandbox has Blender, Rapier physics, Python, a headless browser to test its own work, and a broad library of materials, 3D models and characters.
Human panelists compare blind pairs of scenes built from the same brief and spend at least 30 seconds with each scene before they vote. Votes are then converted into Elo, with 95% confidence intervals.