MoRI + vLLM: Wide Expert Parallelism and RDMA KV-Cache Transfer for Disaggregated MoE Serving on AMD

Описание

Running large Mixture-of-Experts models across pods creates two major challenges: expert dispatch and combine traffic, and moving the KV cache between prefill and decode. At North America 2026, Ravi Gupta, Rishi Madduri, and Shiksha Patel of AMD will present “MoRI + vLLM: Wide…