GPT-6 Astra Scores 95% on Robot Control, 6.2x Fewer Tokens Than Fable 5.1's 40%

scaling01 · x · 2026-09-05

A thread cited by @chooijeq claims GPT-6 Astra scored 95% on a robot control task versus Fable 5.1's 40%, while using 6.2x fewer output tokens at 2.3x lower cost. @scaling01 amplifies it with a bold call that GPT-7 will automate all blue-collar work. Third-party, unverified numbers.

Original post →

More from Embodied

Embodied channel →