G0.5: One Autoregressive Stream for Robot Reasoning and Action
G0.5 is a pretrained autoregressive VLA in which a single transformer decoder emits reasoning and action tokens under one objective. A cross-embodiment action tokenizer, a native chain-of-thought stream, and a visual memory module make this tractable at foundation-model scale, letting the pretrained VLM capabilities carry over to physical behavior. It surpasses pi-0.5 and GR00T-N1.7 across 7 regimes including real-world R1 robot fine-tuning, BEHAVIOR Challenge, DROID zero-shot transfer, LIBERO, RoboTwin 2.0 and SimplerEnv.
Yicheng Liu, Zibin Dong, Baijun Ye·Aug 12, 2026
VLAGalaxeaFoundation ModelsAug 12, 2026
0:09
