LMSYS：Blog（Chatbot Arena 团队）·· 13 小时前AI 评分44
SGLang 用 Waterfill 与 LPLB 改进 DeepEP MoE 负载均衡
Blog Improving DeepEP MoE Load Balance in SGLang with Waterfill and LPLB Mixture-of-Experts (MoE) models rely on Expert Parallelism (EP) to scale inference across multiple GPUs. In SGLang, DeepEP and EPLB provide high-performance serving under EP, but the workload seen by ... NVIDIA Team June 26, 2026
AI 导读
SGLang 为 DeepEP MoE 推理引入 Waterfill 和 LPLB 两项调度期负载均衡特性。
来源：LMSYS：Blog（Chatbot Arena 团队） · lmsys.org