# V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning

*Source* · `src-assran-2025-v-jepa-2` · status: **n/a**

| field | value |
|---|---|
| kind | Preprint |
| authors | Mido Assran, Nicolas Ballas, Yann LeCun, Michael Rabbat, and others |
| date | 2025-06 |
| url | https://arxiv.org/abs/2506.09985 |
| arxiv | 2506.09985 |
| accessed | 2026-10-24 |

## Notes

Over 1M hours of video and images; V-JEPA 2-AC post-trained on under 62 h of DROID robot video; zero-shot pick-and-place with image goals on Franka arms in two labs; SSv2 77.3, EK-100 R@5 39.7.

## Backlinks (computed)

**cited by:** [Physical-world foundation models](../technique/physical-world-foundation-models.md)

## Citation

Mido Assran, Nicolas Ballas, Yann LeCun, Michael Rabbat, and others (2025-06). *V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning*.  https://arxiv.org/abs/2506.09985

---
Provenance: created in run `2026-10-24-techniques-world-models`.