TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Power of local LLMs especially with new Apple chips

38 点作者 mobilio超过 1 年前

2 条评论

Obertr超过 1 年前
Hey! I am the @karmedge who wrote this post on Twitter<p>I wonder what do you guys think about tool which will be able to automate actions on computer? Like clicking smth, automatically opening the window for me after I type in some request?<p>Something like hyperwriter assistant but for Mac. Very similar to open interpret but with accessibility option<p>I wonder what else is there in the space?
评论 #38386854 未加载
danny_codes超过 1 年前
An important missing piece in the price analysis is that the M3 chip is not designed for continuous load. It&#x27;s the difference between retail and datacenter&#x2F;commercial hardware. You can get an RTX-4090 for 1500. But if you try to train on it for ~a few months it will almost certainly die. There&#x27;s a reason datacenter hardware has a 5-10x price premium.
评论 #38419634 未加载