[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-99b8aabd-0fbe-428f-8f44-4cd2ab5b8066":3},{"id":4,"title":5,"summary":6,"original_url":7,"source_id":8,"tags":9,"published_at":23,"created_at":24,"modified_at":25,"is_published":26,"publish_type":27,"image_url":13,"view_count":28},"99b8aabd-0fbe-428f-8f44-4cd2ab5b8066","Google DeepMind发布Decoupled DiLoCo：跨数据中心分布式训练的新突破","# Google DeepMind发布Decoupled DiLoCo：跨数据中心分布式训练的新突破\n\nGoogle DeepMind于4月23日发布了Decoupled DiLoCo，这是一项结合Pathways异步编排系统与DiLoCo低通信训练方法的分布式训练架构创新，旨在解决超大规模AI模型跨地理分布训练的核心瓶颈。\n\nDecoupled DiLoCo的核心设计思路是\"解耦\"。Pathways系统负责协调异构芯片以独立速度运行，而DiLoCo专注于最小化跨数据中心通信开销。两者结合后，内层优化可在本地完成，外层更新仅进行低频同步，将跨站点通信量降低至原来的八分之一。初步基准测试显示，在分布式设置中可减少高达50%的训练时间。\n\n这一技术突破对当前大模型训练面临的现实挑战具有直接意义。随着模型规模突破万亿参数量级，跨数据中心的互连带宽和\"掉队者效应\"成为训练效率的主要瓶颈。Decoupled DiLoCo通过异步协调和极低带宽需求，使得地理分散的硬件资源能够高效协作训练同一个模型。\n\n从技术生态角度看，该架构支持GPU、TPU甚至边缘设备的混合部署，无需频繁数据交换。这为数据主权合规场景（如GDPR要求下的本地化处理）提供了可行的技术路径，同时也为算力资源不足的地区参与前沿模型训练降低了门槛。\n\n分布式训练效率的提升将直接影响大模型的训练成本和迭代速度，这一方向的持续创新对整个AI行业的基础设施建设至关重要。","https:\u002F\u002Fx.com\u002FGoogleDeepMind\u002Fstatus\u002F2047330987353239925","4d11edad-2df6-45f6-b71f-70f65de7f7fd",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"40269b40-7942-4650-9672-ed2e6524d37a","ai-technology",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"8cf7490f-2449-4ba7-be19-61befa0d92b4","google",{"id":18,"name":19,"slug":19,"description":13,"color":13},"0a93ec8e-ea39-4693-81de-563ca8c173f7","inference",{"id":21,"name":22,"slug":22,"description":13,"color":13},"01598627-1ea6-4b27-a5d8-874971571a71","llm","2026-04-24T02:30:00Z","2026-04-23T22:07:46.541869Z","2026-04-23T22:07:46.541884Z",true,"agent",5]