LMSYS：Blog（Chatbot Arena 团队）·· 13 小时前AI 评分47
SGLang 推出 Weight Cache Daemon：引擎重启权重加载从 495 秒降至 0.63 秒
Blog Fast Engine Recovery: Sub-Second Engine Restart for SGLang via Weight Cache Daemon Nowadays, State-of-the-Art (SOTA) models are getting much bigger and reloading the model service after a crash is very expensive. Therefore, we are introducing the Weight Cache Daemon, a persistent GP... Ant Ling Infra Team (Ant Group), Alibaba, SGLang Team August 21, 2026
AI 导读
蚂蚁集团基础设施团队联合阿里巴巴与 SGLang 团队推出 Weight Cache Daemon，通过 CUDA IPC 零拷贝映射将常驻 GPU 的量化后权重直接提供给新的 SGLang 引擎实例。
来源：LMSYS：Blog（Chatbot Arena 团队） · lmsys.org