<?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>VRAM 32GB on Negi AI Lab</title><link>https://ai.negi-lab.com/tags/vram-32gb/</link><description>Recent content in VRAM 32GB 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>Mon, 22 Jun 2026 03:21:06 +0900</lastBuildDate><atom:link href="https://ai.negi-lab.com/tags/vram-32gb/index.xml" rel="self" type="application/rss+xml"/><item><title>ローカルLLMにRTX 5090は必要か？4090比較と失敗しない選び方ガイド</title><link>https://ai.negi-lab.com/posts/rtx-5090-vs-4090-local-llm-guide/</link><pubDate>Mon, 22 Jun 2026 00:00:00 +0900</pubDate><guid>https://ai.negi-lab.com/posts/rtx-5090-vs-4090-local-llm-guide/</guid><description>Llama 3 70Bを実用速度（15 tokens/sec〜）かつ高精度で動かすなら、RTX 5090のVRAM 32GB（想定）が唯一の選択肢になる。...</description></item></channel></rss>