<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title>推理架构 - 标签 -</title><link>https://oklife.me/tags/%E6%8E%A8%E7%90%86%E6%9E%B6%E6%9E%84/</link><description>推理架构 - 标签 -</description><generator>Hugo -- gohugo.io</generator><language>zh-CN</language><managingEditor>contact@oklife.me (梦行志)</managingEditor><webMaster>contact@oklife.me (梦行志)</webMaster><copyright>© 2026 梦行志 | 保留所有权利</copyright><lastBuildDate>Wed, 21 Oct 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://oklife.me/tags/%E6%8E%A8%E7%90%86%E6%9E%B6%E6%9E%84/" rel="self" type="application/rss+xml"/><item><title>12G 本地推理生产线 ⑦ 接入 Hermes 的零成本推理架构：本地主力 + 云端兜底</title><link>https://oklife.me/2026/10/21/sc117-series-06-hermes-wiring/</link><pubDate>Wed, 21 Oct 2026 00:00:00 +0000</pubDate><author>contact@oklife.me (梦行志)</author><guid>https://oklife.me/2026/10/21/sc117-series-06-hermes-wiring/</guid><description>&lt;div class="featured-image">
                &lt;img src="/images/Code-Art-Studio-images/sc117-series-06-hermes-wiring/sc117-series-06-hermes-wiring.webp" referrerpolicy="no-referrer">
            &lt;/div>把 12G 卡上的本地 Strata 模型接进 Hermes Agent 的完整实操：接线配置、双层架构设计、30K prompt / 工具调用 / 长生成实测数据，以及一个隐蔽的 max_tokens 互锁坑。</description></item></channel></rss>