<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[4090 48G最适合跑什么LLM？]]></title><description><![CDATA[<p dir="auto">如果2个4090 48G，能用来做视频生成工具么？</p>
]]></description><link>https://lcz.me/topic/375/4090-48g最适合跑什么llm</link><generator>RSS for Node</generator><lastBuildDate>Wed, 01 Jul 2026 10:02:13 GMT</lastBuildDate><atom:link href="https://lcz.me/topic/375.rss" rel="self" type="application/rss+xml"/><pubDate>Sun, 31 May 2026 22:06:14 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to 4090 48G最适合跑什么LLM？ on Mon, 01 Jun 2026 07:11:51 GMT]]></title><description><![CDATA[<p dir="auto">视频模型，只能单卡跑，不是chat或者多模态模型。</p>
]]></description><link>https://lcz.me/post/4506</link><guid isPermaLink="true">https://lcz.me/post/4506</guid><dc:creator><![CDATA[mei li]]></dc:creator><pubDate>Mon, 01 Jun 2026 07:11:51 GMT</pubDate></item><item><title><![CDATA[Reply to 4090 48G最适合跑什么LLM？ on Mon, 01 Jun 2026 06:57:03 GMT]]></title><description><![CDATA[<p dir="auto">单卡就够了，两张浪费了，你就是4张，也只能跑Qwen3.6 27b，跑Deepseek本地部署意义不大。</p>
]]></description><link>https://lcz.me/post/4505</link><guid isPermaLink="true">https://lcz.me/post/4505</guid><dc:creator><![CDATA[terry]]></dc:creator><pubDate>Mon, 01 Jun 2026 06:57:03 GMT</pubDate></item><item><title><![CDATA[Reply to 4090 48G最适合跑什么LLM？ on Mon, 01 Jun 2026 01:16:18 GMT]]></title><description><![CDATA[<p dir="auto">单4090 48GB也可以跑ComfyUI，可以看一下音视图区他们分享的工作流</p>
]]></description><link>https://lcz.me/post/4487</link><guid isPermaLink="true">https://lcz.me/post/4487</guid><dc:creator><![CDATA[kop wang]]></dc:creator><pubDate>Mon, 01 Jun 2026 01:16:18 GMT</pubDate></item><item><title><![CDATA[Reply to 4090 48G最适合跑什么LLM？ on Mon, 01 Jun 2026 07:53:48 GMT]]></title><description><![CDATA[<blockquote>
<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/huang-michael" aria-label="Profile: Huang-Michael">@<bdi>Huang-Michael</bdi></a> <a href="/post/4479">说</a>:</p>
<p dir="auto">如果2个4090 48G，能用来做视频生成工具么？</p>
</blockquote>
<p dir="auto">视频跟llm 是两套</p>
<p dir="auto">ComfyUI 会需要跑很多不一样的大模型，需要下载很多</p>
<p dir="auto">llm的话个人选择27b dense 模型</p>
<p dir="auto">你可以同时跑一张卡但是会抢算力</p>
<p dir="auto">一张跑comfyUI 一张跑llm<br />
有点奢侈</p>
]]></description><link>https://lcz.me/post/4485</link><guid isPermaLink="true">https://lcz.me/post/4485</guid><dc:creator><![CDATA[applejuice]]></dc:creator><pubDate>Mon, 01 Jun 2026 07:53:48 GMT</pubDate></item></channel></rss>