Google 发布 Gemini Robotics 2.5 物理世界模型
Google 把 Gemini 模型搬进了真实工厂，Apollo 2 机器人给奔驰分拣零件准确率 96%。和 Figure、Tesla 自研路线不同，Google 走模型授权模式。做具身智能和工业自动化的团队可以关注它会不会开放 API。
Google 宣布推出 Gemini Robotics 2.5，一个面向物理 AI 场景的机器人模型，底层由 Gemini 4 Argon 模型提供支持。
在与 Apptronik 和 Mercedes-Benz 的概念验证合作中，Gemini Robotics 2.5 已驱动 Apollo 2 机器人在汽车装配场景中拾取和整理零部件，准确率达到 96%。这说明该模型不只是实验室演示，而是进入了真实产线环境的验证阶段。
Google 此前在 2025 年推出 Gemini Robotics 1.5 并强调跨本体泛化能力，2.5 版本延续了从数字世界向物理世界扩展的路线。相比 Tesla Optimus、Figure 等自研端到端方案，Google 选择了模型授权加伙伴落地的方式，通过与 Apptronik 这类本体厂商绑定进入工厂。对做具身智能或工业自动化的开发者来说，这是一个可能开放 API 的机器人基础模型选项。
Google announced a new Gemini Robotics 2.5 model for physical AI.
> This model was built with support from the Gemini 4 Argon model.
> In a POC with Apptronik and Mercedes-Benz, Gemini Robotics 2.5 helped the Apollo 2 robot pick and organize car parts for assembly with 96% accuracy. https://t.co/KQ6Hf6cHKs
This report is categorized under AITOP's "Models" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows and US-China foundation model benchmarks and commercialization comparison. Key entities and technical topics under continuous tracking include Gemini Robotics, Google, Apptronik, 具身智能.
Source attribution: curated from TestingCatalog (@testingcatalog). Cross-source corroboration: 2 independent sources have covered this development (including Google DeepMind), confirming industry-wide signal strength.
- Google DeepMind15:00Source
