Unverified50% confidenceFactExact time
Gemini Robotics 2相比上一代模型在泛化能力上有增强，能对未见过的物体、场景和任务进行更好的迁移和适应
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/31/2026
First Seen
Valid until: 11/29/2026
Sources
Related Entities
Related Claims
UnverifiedGemini Robotics 2的目标是让一个统一的基础模型能够驱动多种形态的机器人完成多样化任务80% similarUnverifiedGemini Robotics系列采用VLA（Vision-Language-Action，视觉-语言-动作）架构，将Gemini的多模态理解能力延伸到物理动作层面69% similarUnverifiedGemini Robotics系列未来可能会发布更多面向研究社区的工具、数据集与基准69% similarUnverifiedGemini 2.5优化了多工具并行调用和嵌套工具调用的准确率67% similarUnverifiedGemini采用Mixture of Experts（MoE，混合专家）架构，允许模型在不成比例增加计算成本的情况下扩展参数量和上下文处理能力66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/829336API
curl https://kongchang.com/api/v1/knowledge/claims/829336MCP
get_claim(id=829336)