Multiagent Systems on Jaehun's Bloghttps://jaehun.me/en/tags/multiagent-systems/Recent content in Multiagent Systems on Jaehun's BlogHugoen-USMon, 05 Oct 2026 09:26:25 +0900Prefill-Free Cross-Family KV Cache Transfer for Heterogeneous Multi-Agent LLMshttps://jaehun.me/en/posts/paper-2609-32259/Mon, 05 Oct 2026 00:00:00 +0900https://jaehun.me/en/posts/paper-2609-32259/<p><a href="https://arxiv.org/abs/2609.32259"target="_blank" class="inline-flex items-center gap-1" >Paper<svg class="h-3 w-3 flex-shrink-0" id="external-link" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 24 24"><path fill="none" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="2" d="M15 3h6v6m-11 5L21 3m-3 10v6a2 2 0 0 1-2 2H5a2 2 0 0 1-2-2V8a2 2 0 0 1 2-2h6"/></svg> </a></p> <h2 id="how-to-hand-off-kv-caches-as-is-between-different-llms-heterofold">How to Hand Off KV Caches As-Is Between Different LLMs: HeteroFold<a href="#how-to-hand-off-kv-caches-as-is-between-different-llms-heterofold" class="heading-anchor" aria-label="Link to this section"></a></h2><p><strong>TL;DR</strong>: HeteroFold, a method that reuses the KV cache between models with different tokenizers and architectures without the receiver re-prefilling, beats existing cache transfer in all six directions and is 10.74× faster than native prefill at 32K context (source: §4.1, Tab.1, Tab.4).</p>