<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>本地AI on 夏洛特的AI实验室</title><link>https://charlotteailab.com/tags/%E6%9C%AC%E5%9C%B0ai/</link><description>Recent content in 本地AI on 夏洛特的AI实验室</description><generator>Hugo</generator><language>zh-CN</language><lastBuildDate>Mon, 01 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://charlotteailab.com/tags/%E6%9C%AC%E5%9C%B0ai/feed.xml" rel="self" type="application/rss+xml"/><item><title>不花一分钱API费，你的MacBook就能跑AI大模型</title><link>https://charlotteailab.com/posts/mac-ai-ollama-mlx/</link><pubDate>Mon, 01 Jun 2026 00:00:00 +0000</pubDate><guid>https://charlotteailab.com/posts/mac-ai-ollama-mlx/</guid><description>Ollama 0.19接入Apple MLX框架，Mac上本地跑AI速度翻倍。32GB内存以上的Mac用户，现在可以零成本跑350亿参数大模型了。</description></item><item><title>不是只有Mac能跑本地AI：AMD用户终于等到了自己的工具</title><link>https://charlotteailab.com/posts/amd-lemonade-ai/</link><pubDate>Mon, 01 Jun 2026 00:00:00 +0000</pubDate><guid>https://charlotteailab.com/posts/amd-lemonade-ai/</guid><description>AMD推出开源本地AI服务器Lemonade，利用NPU+GPU双加速运行大模型。免费、离线、隐私保护，AMD电脑用户的本地AI时代来了。</description></item><item><title>谷歌Gemma 4来了：手机能跑的AI，性能却让人吃惊</title><link>https://charlotteailab.com/posts/gemma4/</link><pubDate>Mon, 01 Jun 2026 00:00:00 +0000</pubDate><guid>https://charlotteailab.com/posts/gemma4/</guid><description>谷歌发布Gemma 4系列开源模型，四个版本从2B到31B，全部Apache 2.0商用开源。最小版本手机能跑，最大版本AIME数学测试89.2%。但中国模型真的落后了吗？</description></item></channel></rss>