Story

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

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

CryptoTokenPostKorean, translatedFiled 39 min ago

GPT-6.2 Astra ranks first with a robot arm evaluation score of 79 points and 668 commands

Original headlineGPT-6.2 아스트라, 로봇팔 평가 79점·668회 명령으로 1위

Open the sourceReported by TokenPost, tokenpost.kr, 20 November 2026, by 최윤서

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

GPT-6.2 Astra topped the AA-Robotics v0.1 robotic-arm control benchmark with an average score of 79, completing 6 of 8 tasks using 668 total arm commands—the fewest among evaluated models. GPT-6.2 Sol scored 78 and Claude Fable 5.2 scored 75, according to Crypto Briefing's report on the 20th. Astra's command efficiency stood out particularly in the box-sorting task, where it used just 92 commands—roughly 75% fewer than Claude Fable 5.2. The results matter because command count, alongside task completion, signals how efficiently models translate instructions into physical robot actions. What to watch next is whether Astra's efficiency advantage holds across harder tasks and broader robotic hardware.

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, Claude Fable 5.2, GPT-6.2 Sol, AA-Robotics v0.1, Crypto Briefing