Zhiyuan Releases Embodied Model GO-3 Preview: Can Make Jianbing Guozi, Work at C

Zhiyuan released the embodied model GO-3 Preview, which uses WRAM architecture to achieve full-chain scalable learning and topped three major industry benchmarks: RoboDojo, RoboTwin, and RoboColiseum. The model supports zero-shot and few-shot tasks, enabling autonomous improvement in scenarios such as convenience stores, and achieving continuous self-evolution.

On October 29, 2026, Zhiyuan officially released the GO-3 Preview, an embodied model that achieves full-chain scaling in pre-training, post-training, and deployment. This model excels in three key benchmarks: RoboDojo, RoboTwin, and RoboColiseum. GO-3 Preview uses a novel architecture called WRAM (World Reasoning Action Model) to integrate perception, prediction, and action, enabling robots to learn from video, simulation, and real-world interaction. The model also features Genie Evolver, a hybrid training system that combines real robots, physics simulations, and world models to improve performance and adaptability.

On October 29, 2026, Zhiyuan released the GO-3 Preview, an embodied model based on the native embodied architecture WRAM, achieving unified task reasoning, future prediction, and action generation. GO-3 Preview ranked first in three major rankings: RoboDojo, RoboTwin, and RoboColiseum, and plans to fully open-source by the end of 2026. The model continuously iterates through pre-training, post-training, and deployment, enabling end-to-end scalable learning and enhancing robots' adaptability and execution capabilities in complex tasks.

This article categorizes companies working on embodied intelligence into four types: independent brain companies, integrated hardware-software manufacturers, tech giants and large model firms, and compute and open platform enterprises. It highlights ACE ROBOTICS as a leading independent brain company with its Kairos 3.1 world model achieving top results in multiple benchmarks. The piece also discusses challenges such as data scarcity and simulation-to-reality gaps, and provides a framework for selecting the right company based on technical, open-source, and application criteria.