智能体
x.com ·
@Sundar Pichai (@sundarpichai)
Google 发布面向工作的通用 Gemini agent，公布六项架构原则
Sundar Pichai 在 Gemini at Work 活动上发布新的 Gemini agent，定位为面向工作的统一通用 Agent，可回答问题、处理知识工作、生成图像与媒体、编写并运行代码，并连接个人工作流、记录系统与企业管控。其架构原则包括统一 Agent、跨设备 Web 访问、云端持久执行与统一记忆、多 Agent 编排、上下文理解及多模型调度。
AI推荐理由 Sundar Pichai 在 Gemini at Work 活动上介绍新的 Gemini agent，并列出统一 Agent、跨设备访问、云端持久执行、多 Agent 编排、上下文记忆、多模型调度等架构原则，附官方博客链接。属于一手产品发布信息，架构描述具体，对企业级 Agent 落地有较高参考价值；但均为厂商自述，尚无第三方验证或效果数据。
84
浏览 251,949
点赞 3,028
评论 214
转发 278
正文
Great to be with 700+ @googlecloud customers at the historic NASA Hangar One for Gemini at Work!
Today, we introduced the new Gemini agent, a single, universal agent for work that has all of your business context and answers your questions, handles your knowledge work, creates your images and media, and writes and runs code - all from a single prompt box. It is connected to your personal workflows, your systems of record, and your enterprise controls.
It’s built around a few core architectural principles:
Unified Agent: Gemini can answer questions, do knowledge work and generate code from a single prompt box.
Access: Gemini is web-based and can be accessed from any device and integrated into third-party applications. It can also operate without a dedicated user interface.
Persistent Execution: It runs in the cloud, which means it maintains a single set of memories, context and one personalization graph no matter where you access it.
Multi-Agent Orchestration: Gemini can create sub-agents to tackle multi-step tasks and can also act as a coworker agent with a defined role and its own dedicated identity
Context: Gemini knows your tools, data, and work history and learns how you work the more you use it.
Model Choice Flexibility: It orchestrates across multiple models to deliver optimal quality and lower your costs.
Read more in our blog: https://cloud.google.com/blog/products/ai-machine-learning/welcome-to-gemini-at-work-2026