Runway releases Praxis-1: video-trained world action model tops NVIDIA RoboLab-120
graceisford · x · 2026-11-12
Runway released Praxis-1, a world action model betting that large-scale video data—not gaming or egocentric data—is the most scalable path to capable robots.
- Ranks #1 on NVIDIA's RoboLab-120 leaderboard, ahead of every open-weight VLA including π0.5 and GR00T N1.6
- Adapts quickly to unseen robots: with under 30 minutes of data it passed all 15 simulated trials on Noble Machines' robot
- Weights are available today
Related event: Runway Open-Sources Praxis-1 World Action Model, Tops RoboLab-120(10 posts)→
More from Embodied
- Baidu World 2026 to showcase Apollo Go robotaxi rides and Apollo World physical AI model — Baidu_Inc · 2026-11-12
- NVIDIA's open-source Alpamayo 2 Edge VLA driving model lands in Autoware — drmapavone · 2026-11-12
- A closer look at Astra's robot gripper design (and a joke at its expense) — _Stocko_ · 2026-11-12
- Robot startup passes first presale milestone with 10+ units claimed — R4nsack · 2026-11-12
- Runway launches Praxis-1, a 27B MoE world action model, with a 73ms Fast variant — ChongZzZhang · 2026-11-12
- Dream-1 on bimanual YAM robot folds a towel on voice command at CoRL — DJiafei · 2026-11-12