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

**Published:** October 14, 2026
**Author:** PyTorch
**Source:** YouTube — PyTorch Official
**Language:** en-US
**HTML:** https://aiqazaqstan.com/en/articles/mori-vllm-wide-expert-parallelism-and-rdma-kv-cache-transfer-for-disaggr

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…

> From the original source: **YouTube — PyTorch Official**
> [Read full article at YouTube — PyTorch Official](https://www.youtube.com/watch?v=cVSkL3dI2b8)

**Tags:** #GPU & Суперкомпьютеры #YouTube

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*