GPT-6 Astra tops a 16-benchmark robot index, yet went 0/11 on real plug insertions

AIFlow_ML · x · 2026-11-06

An analysis of the benchmark-vs-real-world gap in robotics: GPT-6 Astra ranks #1 on a 16-benchmark robot index, yet scored 0 of 11 plug insertions on real robots. It cites investor Freda Duan, who after talking to 50 robotics people claimed 'Astra IS robotics' GPT-3 moment.' Practical takeaway: general model as planner, small fast policy as hands, a checker you trust, and pick tasks by type.

Related event: GPT-6 Astra Tops Robot Benchmarks but Fails All 11 Real-World Plug Tasks(2 posts)→

Original post →

More from Embodied

Embodied channel →