
Model family
Z-sight
Participating organizations
Participating organizationsUniversity of Southern California
Participating organizationsMaven Robotics
Participating organizationsMaven RoboticsVersion scope and roles for these associations await verification.
Z-sight is a 3.18-billion-parameter research robot policy that combines vision-language-action modeling with latent world modeling. It jointly denoises future visual tokens and continuous actions, and was evaluated in simulation and on a physical Unitree G1.
Overview
- Model type
- Action policy
- Architecture
- VLA · WAM · Transformer
- Action generation
- Flow matching
- Listed versions
- Z-sight
- Related robots
- 1
Papers & reports (1)
Z-sight: Unifying Vision-Language-Action and Latent World Modeling
Paper available
arXiv: 2611.24940DOI: 10.48550/arxiv.2611.24940
Authors & affiliations (15)
Author order: Original order
- First published
- 2026-11-19
- Latest paper revision
- v1
- Venue
- Not provided
Covered model versions
- z-sight · Z-sight
Z-sight model and author-reported simulation and Unitree G1 evaluations
Official resources
- Z-sight author project and videos ↗
Publisher not provided · Official website
Publication date unknown · Accessed: 2026-11-19
Submit a source or correction