Ropedia W0 world model trained on 10k hours egocentric video runs at 21.3 fps, open-sourced
liuziwei7 · x · 2026-10-26
Ropedia W0 is a single world model that predicts first-person future video for humans and different robots alike, conditioned on camera motion and hand skeletons.
- Pretrained on 10,000 hours of egocentric human video
- Distilled version runs at 21.3 fps on one GPU
- Code and weights are open-sourced
Related event: Ropedia Releases W0, a Unified World Model for Humans and Robots(2 posts)→
More from Embodied
- Hobbyist teaches StackChan robot to read scripts with Grok and Hermes-Agent — alexcovo_eth · 2026-10-26
- 10 wild humanoid robots: ex-Boston Dynamics droid butler, car-lifting bodyguard bot, ladder-climbing G1 — minchoi · 2026-10-26
- KAIST's humanoid v0.9 pushes 35 kg and lifts 20 kg, then moonwalks — minchoi · 2026-10-26
- Echo teaches VLA robots to detect task success and failure without frame-level labels — Yufeng Liu · 2026-10-26
- Alibaba DAMO releases 16B unified embodied model RynnUnify, 51.5% on RoboCasa365 — Minghao Zhu · 2026-10-26
- HoloBoxing teaches humanoid boxing with just 7 minutes of single-GPU training — Xihan Ma · 2026-10-26