<?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[VLLM塞不下模型]]></title><description><![CDATA[<p dir="auto">小弟目前是在Windows環境用ollama跑龍蝦雲端跟本地<br />
原本是想用Qwen3.5雲端幫忙安裝VLLM跑Qwen3.6 27b Q4<br />
再把龍蝦整個移植進去<br />
但是一直回報說32gb沒辦法容納Qwen3.6 27b Q4模型<br />
請問有大哥跑通的嗎<br />
拜託指導一下  感恩~</p>
<p dir="auto">以下是小弟的設備<br />
處理器	12th Gen Intel(R) Core(TM) i7-12700K (3.60 GHz)<br />
已安裝記憶體(RAM)	128 GB (128 GB 可用)<br />
圖形卡	NVIDIA RTX PRO 4500 Blackwell (32 GB)<br />
儲存體	已使用 1.18 TB/2.75 TB</p>
]]></description><link>https://lcz.me/topic/48/vllm塞不下模型</link><generator>RSS for Node</generator><lastBuildDate>Wed, 20 May 2026 07:04:19 GMT</lastBuildDate><atom:link href="https://lcz.me/topic/48.rss" rel="self" type="application/rss+xml"/><pubDate>Thu, 07 May 2026 12:29:37 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to VLLM塞不下模型 on Fri, 08 May 2026 13:59:27 GMT]]></title><description><![CDATA[<p dir="auto">好的  謝謝指導~</p>
]]></description><link>https://lcz.me/post/556</link><guid isPermaLink="true">https://lcz.me/post/556</guid><dc:creator><![CDATA[用測試]]></dc:creator><pubDate>Fri, 08 May 2026 13:59:27 GMT</pubDate></item><item><title><![CDATA[Reply to VLLM塞不下模型 on Fri, 08 May 2026 01:49:59 GMT]]></title><description><![CDATA[<p dir="auto">上下文开太长了，vllm建议用awq，32g够跑了，24g的话只能玩玩gguf</p>
]]></description><link>https://lcz.me/post/507</link><guid isPermaLink="true">https://lcz.me/post/507</guid><dc:creator><![CDATA[zhiqing]]></dc:creator><pubDate>Fri, 08 May 2026 01:49:59 GMT</pubDate></item><item><title><![CDATA[Reply to VLLM塞不下模型 on Thu, 07 May 2026 15:01:16 GMT]]></title><description><![CDATA[<p dir="auto">好的，謝謝大哥 我試看看</p>
]]></description><link>https://lcz.me/post/476</link><guid isPermaLink="true">https://lcz.me/post/476</guid><dc:creator><![CDATA[用測試]]></dc:creator><pubDate>Thu, 07 May 2026 15:01:16 GMT</pubDate></item><item><title><![CDATA[Reply to VLLM塞不下模型 on Thu, 07 May 2026 14:38:38 GMT]]></title><description><![CDATA[<p dir="auto">KV cache 也要吃 VRAM 啊 ，gpu-memory-utilization 要設定夠高，VRAM 不夠  max_model_len 就不能設定太大</p>
]]></description><link>https://lcz.me/post/473</link><guid isPermaLink="true">https://lcz.me/post/473</guid><dc:creator><![CDATA[linax777]]></dc:creator><pubDate>Thu, 07 May 2026 14:38:38 GMT</pubDate></item><item><title><![CDATA[Reply to VLLM塞不下模型 on Thu, 07 May 2026 22:31:19 GMT]]></title><description><![CDATA[<p dir="auto">你要不懂Linux下载一个lmstudio 或者llama.cpp，5090足够驱动模型，龙虾不能装在模型的宿主机，会搞坏环境，你的电脑性能不错，可以装个虚拟机，带UI的ubuntu，把openclaw或者Hermes放进去</p>
]]></description><link>https://lcz.me/post/466</link><guid isPermaLink="true">https://lcz.me/post/466</guid><dc:creator><![CDATA[terry]]></dc:creator><pubDate>Thu, 07 May 2026 22:31:19 GMT</pubDate></item></channel></rss>