AW-0
AW-0Paper availableCode and weights
Default checkpoint · YAM + ego pretrained checkpoint
Participating organizationsNVIDIA
Participating organizationsUC Berkeley
Participating organizationsUniversity of PennsylvaniaVersion scope and roles for these associations await verification.
AW-0 is a joint robotics research model from UC Berkeley, NVIDIA and the University of Pennsylvania. Its Mixture-of-Transformers jointly predicts reusable skill code, future video and robot actions. The team evaluated it in RoboTwin 2.0 and on physical bimanual manipulation tasks.
Paper available
Author order: Original order
Zhuoyang Liu · Code repository
Publication date unknown · Accessed: 2026-11-22
miniFranka · Model card
Publication date unknown · Accessed: 2026-11-22
AW-0 research team · Source
Published: 2026-10-12 · Accessed: 2026-11-22