Story

On air. The free wire runs 15 minutes behind live. See pricing

World, crypto, finance and tech news, scored for impact.

CryptoTokenPostKorean, translatedFiled 40 h ago

GPT-6.2 Astra tops 3D evaluation with 1,318 points... leads Claude Fable 5.2 by 47 points

Original headlineGPT-6.2 아스트라, 3D 평가 1318점 1위…클로드 페이블 5.2 47점 앞서

Open the sourceReported by TokenPost, tokenpost.kr, 17 November 2026, by 정민석

Not investment advice. Muhabbit scores how much a story may matter, not which way to trade.

OpenAI’s GPT-6.2 Astra debuted atop Artificial Analysis’s AA-3D leaderboard with a 1,318 Elo score, 47 points ahead of second-place Claude Fable 5.2, after OpenAI released Astra and sister model GPT-6.2 Sol on the 17th; Sol scored 1,259 Elo in the same ranking. The AA-3D benchmark tasks models with turning written prompts into interactive 3D scenes delivered as web applications, making it a direct test of generative spatial-reasoning capability rather than text or image output. The result matters because it gives OpenAI a measurable lead in 3D content generation, a segment increasingly relevant to gaming, design and immersive web tooling. Watch next for whether Anthropic and Google respond with updated models, and whether AA-3D scores translate into adoption by developers building 3D applications.

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

GPT-6.2 Astra, OpenAI, Claude Fable 5.2, GPT-6.2 Sol, Artificial Analysis