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 textitGigaBrain-0.5M*, a VLA model trained via world model-based reinforcement learning. Built upon textitGigaBrain-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. textitGigaBrain-0.5M* further integrates world model-based reinforcement learning via textitRAMP (Reinforcement leArning via world Model-conditioned Policy) to enable robust cross-task adaptation. Empirical results demonstrate that textitRAMP achieves substantial performance gains over the RECAP baseline, yielding improvements of approximately 30% on challenging tasks including textttLaundry Folding, textttBox Packing, and textttEspresso Preparation. Critically, textitGigaBrain-0.5M^* exhibits reliable long-horizon execution, consistently accomplishing complex manipulation tasks without failure as validated by real-world deployment videos on our hrefhttps://gigabrain05m.github.ioproject page.