RoboDojo Puts GPT-6 Astra to the Test: Strong Semantics, Weak Physical Commonsense
ZeYanjie · x · 2026-09-16
The RoboDojo team ran a systematic embodied-agent evaluation of GPT-6 Astra, covering simulation and real-robot tests (benchmarked against GPT-5.5 and DeepSeek-Flash), high-level humanoid control, dexterous piano playing via RoboPianist, and a systematic study of in-context learning (ICL).
Key takeaways:
- GPT-6 Astra shows remarkably strong semantic and spatial understanding plus impressive in-context adaptation;
- Physical commonsense remains a clear bottleneck, revealing a gap between understanding the world and truly reasoning about its physics.
The full report and demos are publicly available.
More from Embodied
- Wearing Whoop and Apple Watch on a single dual-slot band — tinyfool · 2026-11-02
- REA: async brain-cerebellum runtime for robots where policies never wait on the LLM — Pengfei Zhou · 2026-11-02
- Robot Combat Sports Coming to Japan with a Permanent Arena in Tokyo — cixliv · 2026-11-02
- REK robot fighting in the cage: humanoid bot in boxing gloves draws a crowd — broodsugar · 2026-11-02
- iPhone Duo: a folding iPhone mod impresses, but its hinge raises crackgate concerns — rudrank · 2026-11-02
- InflatableBots: University of Calgary's inflatable robots bring dynamic haptics to large-scale VR — lukas_m_ziegler · 2026-11-02