Original Summary

Rummaging through some boxes of old stuff, I found one of my old smart phones. It&#x27;s an Asus Zenfone 4 Max (ZC554KL, Snapdragon 430, 8x Cortex-A53, 3 GB RAM, 32 GB storage) that I bought back in 2017 and so it&#x27;s been sitting in a box for ~8 or 9 years. Still works! So I thought I&#x27;d give it a chance to avoid the garbage by turning it into a tiny AI chatbot server that can connect to any wifi network and immediately make itself available to chat via a web UI on the local network.<p>The phone now boots into postmarketOS, with a small touch screen UI for picking a Wi-Fi network. Once connected, the web UI is started and a chat URL provided (the app sets its hostname to ittybitty so the standard chat URL will be <a href="http:&#x2F;&#x2F;ittybitty.local" rel="nofollow">http:&#x2F;&#x2F;ittybitty.local</a> or just the IP address) and a QR code that will take you to the chat UI.<p>llama.cpp was compiled on the phone itself (the first full build took over an hour). The model is Qwen2.5-0.5B-Instruct at 4-bit (403 MB).<p>It can&#x27;t really <i>do</i> a whole lot other than chat and answer questions using it&#x27;s training data, which is tiny so not super-intelligent on its own, but I added a few helpers to the web page (a calculator, the date&#x2F;time, Wikipedia lookups). This project was mostly just to see if it would work at all, the results were relatively impressive.<p>What I&#x27;m seeing:<p>- ~4.5 tokens&#x2F;sec generating, ~13 tokens&#x2F;sec reading the prompt - llama-server uses about 570 MB of RAM with the model loaded - two people can chat at once; users wait their turn - I tried SmolLM2-360M first. It was faster (6 tok&#x2F;s) but it looped and refused to answer things. The 0.5B model is noticeably better at holding a conversation.<p>The coolest thing is it&#x27;s portability. Anywhere you go that has a wifi to connect to you can just turn the phone on, connect it to the wifi, and now that network has a little AI chat bot to talk to. If this old crappy smart phone does this well I can only imagine how newer, better, faster, bigger phones would do.<p>If anybody has this same smart phone and is interested in doing the same thing just let me know and I can give you the custom OS and llamacpp builds to start from.<p>Not a whole lot to &#x27;show&#x27; since it runs inside my local network but here are some screenshots of the mobile app&#x2F;server and the web UI:<p><a href="https:&#x2F;&#x2F;spottedmarley.com&#x2F;ittybitty-home.png" rel="nofollow">https:&#x2F;&#x2F;spottedmarley.com&#x2F;ittybitty-home.png</a><p><a href="https:&#x2F;&#x2F;spottedmarley.com&#x2F;ittybitty-ui.jpg" rel="nofollow">https:&#x2F;&#x2F;spottedmarley.com&#x2F;ittybitty-ui.jpg</a>


  • 情报分类:商业与市场研究
  • 分类依据:内容涉及商业、投资或市场动态
  • 信息来源:Hacker News 新项目
  • 发布时间:2026/9/25 02:42:53