<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Kimi K3 推論速度 on Negi AI Lab</title><link>https://ai.negi-lab.com/tags/kimi-k3-%E6%8E%A8%E8%AB%96%E9%80%9F%E5%BA%A6/</link><description>Recent content in Kimi K3 推論速度 on Negi AI Lab</description><image><title>Negi AI Lab</title><url>https://ai.negi-lab.com/images/og-default.png</url><link>https://ai.negi-lab.com/images/og-default.png</link></image><generator>Hugo -- 0.154.5</generator><language>ja</language><lastBuildDate>Thu, 06 Aug 2026 07:18:17 +0900</lastBuildDate><atom:link href="https://ai.negi-lab.com/tags/kimi-k3-%E6%8E%A8%E8%AB%96%E9%80%9F%E5%BA%A6/index.xml" rel="self" type="application/rss+xml"/><item><title>ローカルLLM環境の選び方と比較：Kimi K3級を動かすRTX・MacのVRAM基準</title><link>https://ai.negi-lab.com/posts/local-llm-gpu-comparison-rtx-mac-vram/</link><pubDate>Thu, 06 Aug 2026 00:00:00 +0900</pubDate><guid>https://ai.negi-lab.com/posts/local-llm-gpu-comparison-rtx-mac-vram/</guid><description>Kimi K3級の超巨大モデルは個人ではクラウド推論一択ですが、実務でのコード生成やRAG開発にはVRAM 24GBのRTX 4090、あるいは統一メモリ...</description></item></channel></rss>