Artificial Analysis Launches AA-3D Benchmark; GPT-6.2 Astra Tops List
Artificial Analysis released its AA-3D v1.0 interactive 3D benchmark, where agents build three.js scenes offline from text briefs; GPT-6.2 Astra topped the leaderboard with 1318 points, and both Claude and GPT recreated a GTA VI-style Miami in one prompt.
2026-11-18 ~ 2026-11-18 · 2 related posts
- Claude and GPT models rebuild Miami as Vice City in 3D, one prompt each — ArtificialAnlys · 2026-11-18
- AA-3D benchmark: GPT-6.2 Astra tops 3D scene building at 1318 Elo — ArtificialAnlys · 2026-11-18