On May 11, 2026, the top story on Hacker News was an essay titled "Local AI needs to be the norm" . 1,646 points. 643 comments. The fifth-ranked story the same day was a practitioner walkthrough — "Running local models on an M4 with 24GB memory" — and its top-rated reply called Gemma 4 31B "the new baseline… less like a science experiment than any previous local model." At #11 on GitHub trending: jundot/omlx , a Mac inference server managed entirely from the menu bar. 13,600 stars. +455 in a day. 📖 Read the full version with charts and embedded sources on ComputeLeap → Three independent signals, same news cycle, same thesis. The frame around local AI has changed. The question used to be "can you run it locally?" — and the answer was a hobbyist's hedged yes. The question this week is "why isn't local the default?" — and the answer comes packaged as a polished menu-bar app running a 31-billion-parameter open model on a $1,599 laptop.…