<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0">
    <channel>
        <title>蒸馏 - 标签 - vllbc02&#39;s blogs</title>
        <link>https://blog.vllbc.top/tags/%E8%92%B8%E9%A6%8F/</link>
        <description>蒸馏 - 标签 - vllbc02&#39;s blogs</description>
        <generator>Hugo -- gohugo.io</generator><language>zh-CN</language><copyright>This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.</copyright><lastBuildDate>Fri, 05 Jun 2026 00:00:00 &#43;0000</lastBuildDate><atom:link href="https://blog.vllbc.top/tags/%E8%92%B8%E9%A6%8F/" rel="self" type="application/rss+xml" /><item>
    <title>Qwen2.5-3.7 系列最全总结</title>
    <link>https://blog.vllbc.top/qwen/</link>
    <pubDate>Fri, 05 Jun 2026 00:00:00 &#43;0000</pubDate>
    <author>vllbc</author>
    <guid>https://blog.vllbc.top/qwen/</guid>
    <description><![CDATA[<blockquote>
<p>来源：<a
href="https://mp.weixin.qq.com/s/vKptJ55Lgr_c6Eppv9BHKA">Qwen2.5-3.7系列最全总结~</a>（作者：自动学习机
/ 丁师兄，知乎原文 <a
href="https://zhuanlan.zhihu.com/p/2045665300545262204">https://zhuanlan.zhihu.com/p/2045665300545262204</a>）</p>
<p>注：Qwen 3.5 之后没有 technical report，只发了
blog，本文很多地方只能基于假设与论文缺省。</p>
</blockquote>
<p>下文按”架构 → 预训练 → 后训练”的主线，整理从 Qwen2.5 到 Qwen3.7
的演进。</p>]]></description>
</item>
</channel>
</rss>
