1st in AA-3D with 1271 points… Claude Fable 5.2
Original headline1271점으로 AA-3D 1위…클로드 페이블 5.2
Not investment advice. Muhabbit scores how much a story may matter, not which way to trade.
Claude Fable 5.2 topped Artificial Analysis's AA-3D leaderboard with a 1271±25 Elo score under the "Max·default fallback" setting, 67 points ahead of the second-place Claude Sonnet 5.5 (1204±28). Claude Opus 5.5 followed at 1202±28, with GPT-6 Astra at 1168±28 and Gemini 4 Argon at 1147±30. The AA-3D benchmark tasks AI models with building interactive 3D scenes as webpages from written prompts, testing spatial and code-generation capability beyond text and image outputs. The result gives Anthropic's Claude family a sweep of the top three positions, reinforcing its lead in agentic, render-ready generation as rivals prepare their next model releases.
Muhabbit's AI wrote this summary in its own words from TokenPost's report and translated it from Korean. It can get details wrong, so check the source.
Named in this story
Claude Fable 5.2, Claude Sonnet 5.5, GPT-6 Astra, Claude Opus 5.5, Gemini 4 Argon, Artificial Analysis