Natura-1: a physical-world foundation model trained on ~1M hours of egocentric video
lukas_m_ziegler · x · 2026-10-13
naturewill released Natura-1, a general foundation model for the physical world trained on nearly one million hours of egocentric human video, combining dexterous control, zero-shot instruction following, and long-horizon execution. The author highlights long-horizon execution as the hardest and least-demoed capability: task state comes from the task's own history, requiring planning, memory and recovery rather than better grasping.
Related event: Nature Will Unveils Natura-1, a Foundation Model for the Physical World(9 posts)→
More from Embodied
- Meta researcher recruiting multimodal and robotics interns for next summer, no background needed — liuzhuang1234 · 2026-10-14
- UNITAS: a 1.7B 3D-native world action model hits 99.8% on LIBERO and 85% on real tasks — CUHKSZ · 2026-10-14
- Frontier LLM scores on real robot benchmark double every 6 weeks, 15% to 48% in 3 months — ycombinator · 2026-10-14
- Raven Prism AI glasses launch set for October 28 in San Francisco — smtabatabaie · 2026-10-14
- DeLorean joins Airspeeder International Series, first American team debuting in November — evilrabbit_ · 2026-10-14
- Gemini 4 Argon taught Apptronik's Apollo humanoid cornhole from video, sinking 7 straight throws — GoogleDeepMind · 2026-10-14