<?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[【ADM显卡优化】Strata v0.1.34 Windows 系统下的AMD 显卡一键启动]]></title><description><![CDATA[<p dir="auto">Windows 系统下的 AMD 显卡：在 Windows 系统下，AMD 显卡的设置方式与 NVIDIA 显卡类似：双击即可START-HERE.bat。<br />
在没有 NVIDIA 显卡的 PC 上，Strata 会自动识别 AMD 显卡并下载现成的 AMD 引擎（strata-windows-x64-hip.zip）。该引擎包含所需的 ROCm 库，因此只需安装 AMD 驱动程序（AMD 软件：Adrenalin Edition）即可。支持的显卡：RX 7900 XT / XTX、RX 7800 XT /<br />
7700 XT、RX 9060 XT、RX 9070 / 9070 XT、Radeon AI PRO R9700、RX 6800 / 6900 系列。<br />
<a href="https://github.com/Niko1221/Strata/releases#release-v0.1.34" rel="nofollow ugc">https://github.com/Niko1221/Strata/releases#release-v0.1.34</a></p>
<p dir="auto"><strong>安装手顺：</strong></p>
<ul>
<li>使用 AI助手安装<br />
使用 AI 助手（例如 Hermess、 Deepseek harness、Claude Code、Cursor、Codex、GitHub Copilot 等）请将以下内容粘贴到其中：</li>
</ul>
<pre><code>Set up Strata on this PC for me: https://github.com/Niko1221/Strata - follow docs/AI_SETUP.md in that repository.
</code></pre>
<ul>
<li>手动安装<br />
下载 <a href="https://github.com/Niko1221/Strata/archive/refs/heads/main.zip" rel="nofollow ugc">Strata压缩包</a>并解压缩（或git clone直接解压）。<br />
Windows 系统：双击运行START-HERE.bat。<br />
Linux系统：./setup.sh在 Strata 文件夹中运行。</li>
</ul>
]]></description><link>https://lcz.me/topic/2041</link><generator>RSS for Node</generator><lastBuildDate>Mon, 12 Oct 2026 15:39:14 GMT</lastBuildDate><atom:link href="https://lcz.me/topic/2041.rss" rel="self" type="application/rss+xml"/><pubDate>Fri, 02 Oct 2026 02:17:07 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to 【ADM显卡优化】Strata v0.1.34 Windows 系统下的AMD 显卡一键启动 on Sun, 04 Oct 2026 00:45:32 GMT]]></title><description><![CDATA[<p dir="auto">@dardeaw-fen<br />
6G显卡的用户，基本不可能配32G以上内存，所以实用性不大。<br />
32G内存只能跑Coder<br />
48G内存只能跑到Q2<br />
64G内存以上才能跑所有模型。</p>
]]></description><link>https://lcz.me/post/22174</link><guid isPermaLink="true">https://lcz.me/post/22174</guid><dc:creator><![CDATA[perter]]></dc:creator><pubDate>Sun, 04 Oct 2026 00:45:32 GMT</pubDate></item><item><title><![CDATA[Reply to 【ADM显卡优化】Strata v0.1.34 Windows 系统下的AMD 显卡一键启动 on Sat, 03 Oct 2026 09:49:59 GMT]]></title><description><![CDATA[<p dir="auto">牛b plus了，内存价格又撑住了。</p>
]]></description><link>https://lcz.me/post/22093</link><guid isPermaLink="true">https://lcz.me/post/22093</guid><dc:creator><![CDATA[tom23]]></dc:creator><pubDate>Sat, 03 Oct 2026 09:49:59 GMT</pubDate></item><item><title><![CDATA[Reply to 【ADM显卡优化】Strata v0.1.34 Windows 系统下的AMD 显卡一键启动 on Sat, 03 Oct 2026 07:49:21 GMT]]></title><description><![CDATA[<p dir="auto"><a class="plugin-mentions-user plugin-mentions-a" href="/user/perter" aria-label="Profile: perter">@<bdi>perter</bdi></a> 6G顯卡能用....那搞不好3050都還能跑出點東西來了</p>
]]></description><link>https://lcz.me/post/22065</link><guid isPermaLink="true">https://lcz.me/post/22065</guid><dc:creator><![CDATA[dardeaw feng]]></dc:creator><pubDate>Sat, 03 Oct 2026 07:49:21 GMT</pubDate></item><item><title><![CDATA[Reply to 【ADM显卡优化】Strata v0.1.34 Windows 系统下的AMD 显卡一键启动 on Sat, 03 Oct 2026 05:22:52 GMT]]></title><description><![CDATA[<p dir="auto">Strata v0.1.38速度提升很大，5070显卡，IQ3_XXS模型，128K上下文，decode 最快速度超过 80+t/s<br />
<img src="https://upload.lcz.me/uploads/73558be4-802e-4460-a53e-a79d8ba280a1.jpeg" alt="image.jpeg" class="img-markdown  img-fluid" /></p>
]]></description><link>https://lcz.me/post/22055</link><guid isPermaLink="true">https://lcz.me/post/22055</guid><dc:creator><![CDATA[perter]]></dc:creator><pubDate>Sat, 03 Oct 2026 05:22:52 GMT</pubDate></item><item><title><![CDATA[Reply to 【ADM显卡优化】Strata v0.1.34 Windows 系统下的AMD 显卡一键启动 on Sat, 03 Oct 2026 03:52:27 GMT]]></title><description><![CDATA[<p dir="auto">Strata v0.1.38 最新更新，主要是对--kv q4_0的优化，以及6G显存显卡的支持。</p>
<ul>
<li>升级安装<br />
<strong>1.下载升级文件</strong><br />
<a href="https://github.com/Niko1221/Strata/releases#release-v0.1.38" rel="nofollow ugc">https://github.com/Niko1221/Strata/releases#release-v0.1.38</a><br />
N卡下载 strata-windows-x64.zip压缩包并解压缩。<br />
A卡下载 strata-windows-x64-hip.zip压缩包并解压缩。<br />
<strong>2.覆盖\Strata\engine目录下所有文件完成升级。</strong><br />
<strong>3.Windows 系统：strata目录双击运行START-HERE.bat运行。</strong></li>
</ul>
]]></description><link>https://lcz.me/post/22036</link><guid isPermaLink="true">https://lcz.me/post/22036</guid><dc:creator><![CDATA[perter]]></dc:creator><pubDate>Sat, 03 Oct 2026 03:52:27 GMT</pubDate></item><item><title><![CDATA[Reply to 【ADM显卡优化】Strata v0.1.34 Windows 系统下的AMD 显卡一键启动 on Fri, 02 Oct 2026 08:12:09 GMT]]></title><description><![CDATA[<p dir="auto">这效果不错</p>
]]></description><link>https://lcz.me/post/21936</link><guid isPermaLink="true">https://lcz.me/post/21936</guid><dc:creator><![CDATA[terry]]></dc:creator><pubDate>Fri, 02 Oct 2026 08:12:09 GMT</pubDate></item><item><title><![CDATA[Reply to 【ADM显卡优化】Strata v0.1.34 Windows 系统下的AMD 显卡一键启动 on Fri, 02 Oct 2026 08:02:07 GMT]]></title><description><![CDATA[<p dir="auto">Qwen3.8-Flash-Next_IQ3_XXS模型能力测试，一次生成。</p>
<p dir="auto"><img src="https://upload.lcz.me/uploads/6ed442c5-af85-43af-b7cd-c5bd1fe6bcfc.jpeg" alt="image.jpeg" class="img-markdown  img-fluid" /><br />
<a href="https://upload.lcz.me/uploads/b290860c-cc6a-4577-84a3-31e8c393fd85.html" rel="nofollow ugc">taiji-bagua.html</a></p>
]]></description><link>https://lcz.me/post/21933</link><guid isPermaLink="true">https://lcz.me/post/21933</guid><dc:creator><![CDATA[perter]]></dc:creator><pubDate>Fri, 02 Oct 2026 08:02:07 GMT</pubDate></item><item><title><![CDATA[Reply to 【ADM显卡优化】Strata v0.1.34 Windows 系统下的AMD 显卡一键启动 on Fri, 02 Oct 2026 04:40:33 GMT]]></title><description><![CDATA[<p dir="auto">Strata 可以在普通 PC 上运行Qwen3.8-Flash-Next——一个通常需要服务器的大型智能 AI 模型。它可以聊天、编写代码、读取图片，并与您的应用程序和编码代理协同工作，所有数据都不会离开您的 PC。<br />
两台普通游戏电脑性能参考。<br />
<img src="https://upload.lcz.me/uploads/a5ff9eb9-9bff-40e7-9c4a-a589a0b62024.jpeg" alt="image.jpeg" class="img-markdown  img-fluid" /></p>
]]></description><link>https://lcz.me/post/21916</link><guid isPermaLink="true">https://lcz.me/post/21916</guid><dc:creator><![CDATA[perter]]></dc:creator><pubDate>Fri, 02 Oct 2026 04:40:33 GMT</pubDate></item></channel></rss>