Physical Agent Capability Index: Astra leads Argon 1349 to 1324 Elo, both fail 4-bottle task
qinzytech · x · 2026-10-23
Embodied Analysis' Physical Agent Capability Index shows Astra leading Argon with 1349 Elo versus 1324.
Both models got the same starting layout and instruction: put all four bottles into the bin, using handovers if needed. Gemini 4 Argon clears the table—its white bottle drops back but it retries and releases it over the bin, with its 69-second recording ending as the final bottle falls. GPT 6 Astra releases the pink and white bottles but repeatedly fails to dispose of the green one, leaving yellow and green on the table when its 99.5-second recording ends, despite being preferred in six of nine direct comparisons in the Oct 20 annotation batch.
Related event: Astra tops embodied capability index but fails bottle task(2 posts)→
More from Embodied
- Robot arm Podrick learns to load utensils by distilling lessons from its own failed runs — ycombinator · 2026-10-23
- Robotics data is far from saturated: labs are hedge funds hunting for data alpha — paigeinsf · 2026-10-23
- How to evaluate robotics data startups: quality annotation and real-world evals win — paigeinsf · 2026-10-23
- Beijing's service robot output surges 170% in first three quarters — yogthos · 2026-10-23
- ETH Zurich robot learns humanoid mobility from 13 minutes of human motion, no reference at deployment — ChongZzZhang · 2026-10-23
- NVIDIA investigates stuttering issues reported on Surface Laptop Ultra — tomwarren · 2026-10-23