<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>KV Cache on KAD</title>
    <link>https://www.kad8.com/tags/kv-cache/</link>
    <description>Recent content in KV Cache on KAD</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>en</language>
    <copyright>© 2026 </copyright>
    <lastBuildDate>Mon, 17 Aug 2026 22:18:05 +0800</lastBuildDate><atom:link href="https://www.kad8.com/tags/kv-cache/index.xml" rel="self" type="application/rss+xml" />
    
    <item>
      <title>NVIDIA BlueField-4 CMX Targets Long-Context AI Inference</title>
      <link>https://www.kad8.com/ai/nvidia-bluefield-4-cmx-targets-long-context-ai-inference/</link>
      <pubDate>Mon, 17 Aug 2026 22:18:05 +0800</pubDate>
      
      <guid>https://www.kad8.com/ai/nvidia-bluefield-4-cmx-targets-long-context-ai-inference/</guid>
      <description></description>
      <media:content xmlns:media="http://search.yahoo.com/mrss/" url="https://www.kad8.com/ai/nvidia-bluefield-4-cmx-targets-long-context-ai-inference/featured-BlueField-4_CMX_data_architecture.jpeg" />
    </item>
    
    <item>
      <title>Marvell CXL Switch Enables 48TB AI Memory Pools</title>
      <link>https://www.kad8.com/hardware/marvell-cxl-switch-enables-48tb-ai-memory-pools/</link>
      <pubDate>Wed, 12 Aug 2026 09:07:14 +0800</pubDate>
      
      <guid>https://www.kad8.com/hardware/marvell-cxl-switch-enables-48tb-ai-memory-pools/</guid>
      <description></description>
      <media:content xmlns:media="http://search.yahoo.com/mrss/" url="https://www.kad8.com/hardware/marvell-cxl-switch-enables-48tb-ai-memory-pools/featured-Marvell_expands_AI_memory_portfolio.jpeg" />
    </item>
    
    <item>
      <title>Why Memory Bandwidth, Not Compute, Determines LLM Inference Speed</title>
      <link>https://www.kad8.com/ai/why-memory-bandwidth-not-compute-determines-llm-inference-speed/</link>
      <pubDate>Sat, 16 May 2026 12:53:51 +0800</pubDate>
      
      <guid>https://www.kad8.com/ai/why-memory-bandwidth-not-compute-determines-llm-inference-speed/</guid>
      <description></description>
      <media:content xmlns:media="http://search.yahoo.com/mrss/" url="https://www.kad8.com/ai/why-memory-bandwidth-not-compute-determines-llm-inference-speed/featured-Reiner_Pope_explains_LLM_latency.jpeg" />
    </item>
    
  </channel>
</rss>
