# RoboJEPA preprint reports goal-image planning on real robots

_Published Wednesday, October 7, 2026 at 10:48 PM EDT · Robotics, Science, AI · Latest · Tier 2 — Notable_

Researchers released RoboJEPA model checkpoints and training and deployment code alongside a preprint reporting that a robot world model can plan toward a single goal image on real hardware without task-specific training. The model learns to predict future states from data spanning 12 robotic embodiments.

The authors say prediction error follows a second-order power law in compute and correlates strongly with robotic planning performance, offering a proxy for real-robot evaluation. They also report tasks requiring long-horizon planning on physical robots. The released model reaches 8B parameters; the scaling and deployment findings are presented in a preprint.

## Sources

- [arXiv Query: search_query=cat:cs.RO&id_list=&start=0&max_results=30](https://arxiv.org/abs/2610.10515v1)

---
Canonical: https://techandbusiness.org/newswire/FUtJy9YRj6RacwKEvOSCMr
Published: 2026-10-08T02:48:26.519Z
Story chronology: 2026-10-07T17:54:42.000Z
Retrieved: 2026-10-08T05:42:19.358Z
Publisher: Tech & Business (techandbusiness.org)