Google 开源 Agent 编排器 AX
英文标题：🔶 Google's Open Agentic Orchestrator
Google 推出 AX，这是一套面向 Agent 执行场景的开放式、声明式控制平面。其设计针对有状态、突发式且长时间运行的 Agent 工作负载，可在模型响应、工具调用或人工审批期间高效暂停与恢复。AX 将任务、工作空间、网络策略和模型抽象为核心原语，旨在帮助开发者和研究人员大规模运行 Agent 集群，并依赖 Agent Substrate 提供底层能力。
1.0 KB
1.3 KB
中文
关于
源于研究，为生产而构建。
AX 诞生于 Google：Agent 运行时系统研究与前沿计算能力在这里交汇。经过多年构建和运营 Agent 执行引擎的经验，Google 内部多个团队认识到，Agent 工作负载代表着一种全新的计算范式：它们是有状态、突发式、长时间运行的执行体，可能在一分钟内进行高强度计算，随后等待模型响应、工具响应或人工审批。为无状态微服务或可预测批处理任务设计的传统编排器，在需要保持空闲沙箱持续运行时成本高得难以承受，同时又缺乏原生的亚秒级暂停与恢复支持。
依托 Google DeepMind 的 Agent 运行时研究，以及在大规模隔离、恢复和调度方面的深厚经验，AX 正在被构建为一个专为 Agent 执行打造的开放式、声明式控制平面。它将任务、工作空间、网络策略和模型抽象为核心原语，使开发者和研究人员无需重新构建底层基础设施，便可运行大规模 Agent 集群。该项目高度依赖 Agent Substrate，但提供了 Agent 抽象和生成式运行时组件。
English / 原文
About
Born from research, built for production.
AX was born at Google when agentic runtime systems research met frontier compute. Over years of building and operating agentic execution engines, teams across Google recognized that agentic workloads represent an entirely new computing paradigm: stateful, bursty, long-running actors that compute intensely for a minute and then wait for model responses, tool responses, or human approval. Traditional orchestrators built for stateless microservices or predictable batch jobs become cost-prohibitive when keeping idle sandboxes running, yet lack native support for sub-second suspend and resume.
Drawing on agentic runtime research from Google DeepMind alongside deep experience in large-scale isolation, resumption, and scheduling, AX is being built as an open, declarative control plane purpose-built for agent execution. It abstracts tasks, workspaces, network policies, and models into core primitives so developers and researchers can run massive fleets of agents without reinventing the underlying infrastructure. This project heavily relies on Agent Substrate but provides agentic abstractions and generative runtime components.