Original Summary

Hi everyone, I made whatmodel. It&#x27;s a simple site that can be visited in the browser or used from the terminal via a:<p>curl whatmodel.sh<p>To use it, 1. Enter your RAM and GPU 2. See what models you can run locally with its rank and the command to run it 3. Compare them with API models on the same score<p>Building it made me realise a 32GB Mac can now run a model that scores on par with GPT-5, completely locally and for free (e.g. Qwen 3.8 27B at 4-bit).<p>Each model gets one score fitted across 18,731 published benchmark results, and every result links to its source.<p>The method is at <a href="https:&#x2F;&#x2F;whatmodel.sh&#x2F;method" rel="nofollow">https:&#x2F;&#x2F;whatmodel.sh&#x2F;method</a> and the data, scoring code and bench script are open source: <a href="https:&#x2F;&#x2F;github.com&#x2F;jakemanger&#x2F;whatmodel" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;jakemanger&#x2F;whatmodel</a><p>If you&#x27;d like to contribute, there&#x27;s a bench script to collect real-world tok&#x2F;s and you can also submit any new models and benchmark scores along with their source on github.

  • 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：2026/10/12 17:18:42