Researchers: Astra is more 3D-aware than other frontier multimodal models, likely trained on 3D data

georgiagkioxari · x · 2026-09-18

Georgia Gkioxari and Michael Black discuss Google's Astra model's spatial understanding. Black notes Astra isn't "natively" 3D/4D but understands enough to help scale 3D/4D data collection. Gkioxari adds that Astra is more 3D-aware than other frontier multimodal models, and since such capabilities generally come from pretraining or RLVR, 3D data was likely included in the training mix — which she argues is the sensible choice. The exchange touches on where spatial reasoning in multimodal models comes from.

Related event: Researchers Suspect Google Astra Trained on 3D Data for Strong Spatial Understanding(2 posts)→

Original post →

More from Models

Models channel →