<?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>长上下文 on 梦兽编程</title><link>https://rexai.top/tags/%E9%95%BF%E4%B8%8A%E4%B8%8B%E6%96%87/</link><description>Recent content in 长上下文 on 梦兽编程</description><generator>Hugo -- 0.163.3</generator><language>zh-cn</language><copyright>梦兽编程</copyright><lastBuildDate>Fri, 28 Aug 2026 10:00:00 +0800</lastBuildDate><atom:link href="https://rexai.top/tags/%E9%95%BF%E4%B8%8A%E4%B8%8B%E6%96%87/index.xml" rel="self" type="application/rss+xml"/><item><title>大模型的“斩杀线”是谁划定的？别让跑分替代真实工程能力</title><link>https://rexai.top/ai/llm/model-benchmark-engineering-reality/</link><pubDate>Fri, 28 Aug 2026 10:00:00 +0800</pubDate><guid>https://rexai.top/ai/llm/model-benchmark-engineering-reality/</guid><description>从 Prompt Cache、Thinking Token、长上下文到推理系统，重新审视国产大模型的真实成本与工程能力。</description></item></channel></rss>