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。

Show HN: Open Computer Use

9 点作者 mlejva4 个月前
Hi everyone, I&#x27;m Vasek (<a href="https:&#x2F;&#x2F;x.com&#x2F;mlejva" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;mlejva</a>), the CEO of the company behind this - <a href="https:&#x2F;&#x2F;e2b.dev" rel="nofollow">https:&#x2F;&#x2F;e2b.dev</a>. The company is called E2B. We&#x27;re an open-source (<a href="https:&#x2F;&#x2F;github.com&#x2F;e2b-dev">https:&#x2F;&#x2F;github.com&#x2F;e2b-dev</a>) devtool that makes it easy to run untrusted AI-generated code in our secure sandboxes. You can think of us as coding runtime for LLMs. You can self host us on GCP (<a href="https:&#x2F;&#x2F;github.com&#x2F;e2b-dev&#x2F;infra&#x2F;blob&#x2F;main&#x2F;self-host.md">https:&#x2F;&#x2F;github.com&#x2F;e2b-dev&#x2F;infra&#x2F;blob&#x2F;main&#x2F;self-host.md</a>) and we&#x27;re working on AWS, then Azure, and any Linux machine.<p>This repo is one of our open-source projects that we&#x27;re releasing to show developers what they can build with E2B. We used our sandboxes that are powered by AWS&#x27;s Firecracker and gave them GUI with Linux. At the same time we made it easy to control this cloud computer with our Desktop SDK (<a href="https:&#x2F;&#x2F;github.com&#x2F;e2b-dev&#x2F;desktop">https:&#x2F;&#x2F;github.com&#x2F;e2b-dev&#x2F;desktop</a>). Essentially building a virtual desktop computer for AI and we gave it to LLMs to control it. Here&#x27;s a demo we showed at an event - <a href="https:&#x2F;&#x2F;x.com&#x2F;tereza_tizkova&#x2F;status&#x2F;1878834392891838556" rel="nofollow">https:&#x2F;&#x2F;x.com&#x2F;tereza_tizkova&#x2F;status&#x2F;1878834392891838556</a><p>We think computer use is still highly experimental but it feels sort of similar like AI codegen in early 2023. You see the sparks but it&#x27;s not there just yet. However, we wanted to research if open-source LLMs could at least get some results. Here we&#x27;re using Llama 3.2, Llama 3.3, and an OS-Atlas as a base model (fine-tuned Qwen).<p>If you have any questions, happy to answer them!<p>We&#x27;re also hiring! If you&#x27;re a fullstack engineer, distributed systems engineer, AI engineer, product designer, or GTM person and based in SF send me a hello to vasek @ e2b.dev!

1 comment

jakubnovak4 个月前
Can I see what&#x27;s happening inside the computer realtime?