GigaBrain-0.5M*: a VLA That Learns From World Model-Based Reinforcement Learning

Vision-language-action (VLA) models that directly predict multi-step action chunks from current observations face inherent limitations due to constrained scene understanding and weak future anticipation capabilities. In contrast, video world models pre-trained on web-scale video corpora exhibit robust spatiotemporal reasoning and accurate future prediction, making them a natural foundation for enhancing VLA learning. Therefore, we propose $\mathit{\text{GigaBrain-0.5M*}}$, a VLA model trained via world model-based reinforcement learning. Built upon $\mathit{\text{GigaBrain-0.5}}$, which is pre-trained on over 10,000 hours of robotic manipulation data, whose intermediate version currently ranks first on the international RoboChallenge benchmark. $\mathit{\text{GigaBrain-0.5M*}}$ further integrates world model-based reinforcement learning via $\mathit{\text{RAMP}}$ (Reinforcement leArning via world Model-conditioned Policy) to enable robust cross-task adaptation. Empirical results demonstrate that $\mathit{\text{RAMP}}$ achieves substantial performance gains over the RECAP baseline, yielding improvements of approximately 30% on challenging tasks including $\mathtt{\text{Laundry}}\mathtt{\text{Folding}}$, $\mathtt{\text{Box}}\mathtt{\text{Packing}}$, and $\mathtt{\text{Espresso}}\mathtt{\text{Preparation}}$. Critically, $\mathit{\text{GigaBrain-0.5M}}{}^{\ast }$ exhibits reliable long-horizon execution, consistently accomplishing complex manipulation tasks without failure as validated by real-world deployment videos on our \href{https://gigabrain05m.github.io}{project page}.