GPT-6 Astra leads Gemini 3.8 Flash 1,349 to 972 on physical agent Elo index
qinzytech · x · 2026-10-22
embodianalysis reports GPT-6 Astra leads Gemini 3.8 Flash on its Physical Agent Capability Index, 1,349 to 972 Elo. Yet in a simulated kitchen task — place a croissant and a jar of jam on a cutting board — both fell short: Flash placed the croissant but never the jam, while Astra dropped the croissant on the floor and moved on. The authors note a binary success metric would mark both as failures, hiding the gap between partial progress and losing an object; the aggregate index and single-run behavior must be read together.
More from Embodied
- Robotics community racing overnight to adopt new whole-body policy validation paper — carlosdponx · 2026-10-22
- Tesla Q3 earnings: Musk fields 21 questions on Robotaxi costs, FSD revenue and Optimus — XFreeze · 2026-10-22
- Musk: Tesla needs 10,000-20,000 Optimus bots doing self-play to bootstrap training data — XFreeze · 2026-10-22
- Riding the first production Mercedes shipped with NVIDIA's full self-driving stack — HarperSCarroll · 2026-10-22
- Musk: Austin Cybercab fleet grew 10x in a month as autonomous miles scale exponentially — XFreeze · 2026-10-22
- CMU paper defines solver-sufficient symbols to cut hybrid robot planning cost — Yiyuan Pan · 2026-10-22