Zhiyuan Releases GenieStudio 2.0, Robots Have a New Battlefield
智元发布GenieStudio2.0，机器人有了新战场
Zhiyuan released the embodied intelligent operation platform Genie Studio 2.0 during the AI WAVE seven consecutive releases event. The platform covers the full lifecycle of robots from model training to actual deployment, supporting completion of new scenario deployment verification within 7 days. The platform addresses skill training, task orchestration, and multi-robot maintenance through three major modules: Flywheel, Agent, and Monitor, and has achieved high success rates and stable operation in multiple industrial scenarios.
Genie Studio 2.0 represents a shift from lab-based demonstrations to real-world deployment in robotics, emphasizing the ability to integrate models, data, and physical systems into operational workflows. The platform's focus on 'deployment-state' capabilities, such as 7×24-hour stability and multi-robot fleet management, highlights its practical application in industrial settings.
Source: Chinese robotics — Dataset and collection discovery · Read original article ↗
English translation unavailable. Choose the original article to read it.
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
Reported success rate
99%
View original evidence
据官方披露，平台可统一管理真机、仿真及开源数据集，支持图像感知模型、视觉—语言—动作模型（VLA）与真机强化学习模型的训练，并接入GO-2具身基座模型；其自研强化学习训练框架在真实项目验证中实现小时级收敛，任务成功率可达99%。
Open source S10Reported success rate
>99.9%
View original evidence
智元公布的四组落地案例，勾勒出这套系统的实际边界： 常州数控机床项目围绕6台数控机床的上下料需求，引入机器人与自动导引车协同作业，14天完成从需求梳理到产线验证的完整流程； 均普智能BMS装配产线上，三台精灵G2协同完成汽车电池管理系统控制器零部件的上料、装配、入箱与转运，面对高反光金属物体和小于1毫米的放置精度要求，精密上下料与装配任务成功率稳定达到99.9%以上，单次作业节拍控制在13秒以内，并完成7×24小时稳定生产验证； 龙旗科技平板电脑上下料产线实现了多台机器人集群统一运维； 广州地铁站点服务场景则已支持60余台设备规模化部署，通过全局看板集中管理不同站点的设备状态与平均无故障时间。
Open source S19
- paper: not_reported
What remains unknown
Not established in the collected evidence: Environment, Control, Data origin.
Reported performance applies to the described task. It does not establish general autonomy or deployment readiness.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
据官方披露，平台可统一管理真机、仿真及开源数据集，支持图像感知模型、视觉—语言—动作模型（VLA）与真机强化学习模型的训练，并接入GO-2具身基座模型；其自研强化学习训练框架在真实项目验证中实现小时级收敛，任务成功率可达99%。
Open source S10
智元公布的四组落地案例，勾勒出这套系统的实际边界： 常州数控机床项目围绕6台数控机床的上下料需求，引入机器人与自动导引车协同作业，14天完成从需求梳理到产线验证的完整流程； 均普智能BMS装配产线上，三台精灵G2协同完成汽车电池管理系统控制器零部件的上料、装配、入箱与转运，面对高反光金属物体和小于1毫米的放置精度要求，精密上下料与装配任务成功率稳定达到99.9%以上，单次作业节拍控制在13秒以内，并完成7×24小时稳定生产验证； 龙旗科技平板电脑上下料产线实现了多台机器人集群统一运维； 广州地铁站点服务场景则已支持60余台设备规模化部署，通过全局看板集中管理不同站点的设备状态与平均无故障时间。
Open source S19
Source：Chinese robotics — Dataset and collection discovery · feng.ifeng.com