Tencent Robotics X 开源三个具身基础模型:VLM、RxBrain、VLA

Tencent Robotics X Open-Sources Three Embodied Foundation Models: Chief Scientist Zhang Zhengyou Explains the 3-Layer Brain Architecture Solving Robot Reaction Speed

精选理由

腾讯开源了三个具身模型,像搭积木一样分层解决机器人从看懂到动起来的难题,VLA 能做到毫秒级反应,做机器人开发的一定要看看。

AI 摘要

腾讯 Robotics X 在 WAIC 2026 上开源三个具身基础模型:VLM 负责场景理解,RxBrain 认知模型基于视觉状态进行规划,VLA 连续动作控制频率达 500-1000Hz。首席科学家张正友解释了三层大脑架构,解决了机器人反应速度问题。这些模型分别对应感知、认知和动作层,可组合用于复杂操作任务。

原文 · pandaily

Tencent Robotics X Open-Sources Three Embodied Foundation Models: Chief Scientist Zhang Zhengyou Explains the 3-Layer Brain Architecture Solving Robot Reaction Speed

Tencent open-sources three embodied models at WAIC 2026: VLM for scene understanding, RxBrain cognitive model for planning with visual states, and VLA for continuous action at 500-1000Hz.