Microsoft announced Project Zenith today, a distraction-free Windows configuration for developer-class PCs with 64GB or more of unified memory that ships pre-set for coding and can run AI models above 30 billion parameters locally, unmetered.
Read the full story: Microsoft's Project Zenith Reboots Windows for AI Developers →
Transcript
Microsoft just announced Project Zenith, and it changes how Windows sets itself up. Buy the right hardware, and Windows arrives already configured to code. File extensions visible. Hidden files on. VS Code and Terminal pinned to the taskbar. WSL wired in with containers, no setup commands needed. But there's a catch. Zenith needs 64 gigabytes of unified memory and serious memory bandwidth, a bar almost no Windows laptop clears today. The payoff is real. Qualifying hardware runs AI models above 30 billion parameters locally, completely unmetered, no cloud tokens required. Right now exactly one machine qualifies, AMD's Ryzen AI Halo mini desktop, the same lineup AMD used to launch a much bigger workstation the same day. More hardware partners are coming. For now, Windows just split into two experiences, depending on how much memory you can afford.