GPT-6 Astra appears to show a "step change" in spatial reasoning based on early benchmarks
decoder ·
OpenAI's GPT-6 Astra appears to represent a major advance in spatial reasoning based on early benchmark results. A new robotics benchmark called StationeryBench compared GPT-6 Astra with Ai2's MolmoAct2 on five desk-object tasks, including uncapping a marker, pouring paper clips, and passing a ruler between two robot arms. Both models controlled the same dual-arm YAM robots across 200 trials. Astra fully completed 7 out of 100 tasks, while MolmoAct2 completed none. Astra's median progress score was 46 out of 100, compared with 12 for MolmoAct2, and all results, videos, and code are available on GitHub. Cornell and Google DeepMind researcher Yoav Artzi described Astra as a step change in spatial reasoning, and on the still-unpublished REMAP benchmark its accuracy is close to human level, though Artzi noted it still does not match humans in other scenarios. Artzi suspects OpenAI trained the model on large amounts of 3D data such as Blender scenes, which aligns with Astra's particular improvement on 3D tasks; OpenAI also has long-term plans to build its own consumer robots.
decoder