<?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>端侧推理 on AI 实战派 · 从技术到赚钱</title><link>https://guijiagi.com/tags/%E7%AB%AF%E4%BE%A7%E6%8E%A8%E7%90%86/</link><description>Recent content in 端侧推理 on AI 实战派 · 从技术到赚钱</description><generator>Hugo</generator><language>zh-cn</language><copyright>本站内容采用 CC BY-NC-SA 4.0 国际许可协议授权</copyright><lastBuildDate>Thu, 22 Oct 2026 14:06:00 +0800</lastBuildDate><atom:link href="https://guijiagi.com/tags/%E7%AB%AF%E4%BE%A7%E6%8E%A8%E7%90%86/index.xml" rel="self" type="application/rss+xml"/><item><title>「Liquid AI发LFM2.5-24B-A2B」2B激活参数怎么做到端侧智能体？</title><link>https://guijiagi.com/posts/2026-10-22-01-liquid-ai-lfm25-24b-a2b-edge-agent/</link><pubDate>Thu, 22 Oct 2026 14:06:00 +0800</pubDate><guid>https://guijiagi.com/posts/2026-10-22-01-liquid-ai-lfm25-24b-a2b-edge-agent/</guid><description>Liquid AI推出LFM2.5-24B-A2B推理模型，总参数24B但仅2B激活，专为端侧私密智能体设计，在Apple M5 Max上可达184 tok/s，主打低延迟与低成本本地部署。</description></item></channel></rss>