Dyna Robotics' New Model Trained on 1M Hours of Human Video Hits 90% Task Success
Dr_Singularity · x · 2026-08-11
Dyna Robotics has unveiled a new robotics foundation model, the DYNA-2 World-Action Model. The model is trained entirely on over 1 million hours of egocentric human video—equivalent to roughly 170 years of continuous waking experience—rather than traditional robot teleoperation data.
Using a world-modeling architecture that combines next-frame and next-action prediction, the system learns physical skills by observing how humans interact with their environment. In tests, scaling up pre-training increased task success rates in high-precision manufacturing from 20% to 80%-90%. The model also demonstrated strong transferability, successfully applying its knowledge across stationary robot arms, humanoid prototypes, and dexterous robotic hands.
Related event: Dyna-2: Million-Hour Human Video Pretraining Unlocks Robot Scaling Laws(23 posts)→
More from Embodied
- Mistral Ventures into Robotics: Introduces VLM Navigation Research, Hits SOTA on R2RCE — sivareddyg · 2026-08-11
- Enfold: Embeds World Models into Representations, Slashing Robot Control Latency by 10x — Weili Zeng · 2026-08-11
- Samsung Official Details Engineering Secrets of Galaxy Z Fold2 Hinge — MarwaEldiwiny · 2026-08-11
- INTACT by ZJU & Tsinghua: Robots Skip Trial-and-Error to Act Directly on Intent — jiqizhixin · 2026-08-11
- Kyle Vogt's Bot Company Sued for Turning Airbnb into Robot Lab — MarwaEldiwiny · 2026-08-11
- Dyna-2 Model Transfers to WUJI2 Dex Hand, Adapts in 13 Minutes of Data — JasonMa2020 · 2026-08-11