Microsoft has announced Project Zenith, which is a stripped-down version of Windows 11 for AI developers. It's built for devices with at least 64GB of unified memory and 250+ GB/s memory bandwidth, with the goal, as Microsoft puts it, of delivering a "ready-to-code distraction-free Windows experience" on "developer-class" devices.
With that, AMD confirms that its Ryzen AI Halo systems (aka AMD's answer to NVIDIA's DGX Spark mini PC) will be the first to support Project Zenith out of the box, with this customized, developer-focused version of Windows pre-installed with Visual Studio Code, WSL, GitHub Copilot CLI, Python 3.14+, WSL 2+ Ubuntu, PowerShell, and more to help streamline development, testing, and iteration. Microsoft confirms that more devices from its OEM partners will support Project Zenith in the coming months.
Even for enthusiasts who aren't AI developers, some of Project Zenith's out-of-the-box features and settings sound like an improvement over vanilla Windows 11. For example, by default, File Explorer in Project Zenith shows file extensions, hidden files, full paths in the title bar, and details. Also, Start menu tips and account notifications are turned off to "reduce distractions."
As an AI-focused version of Windows, Microsoft notes that Project Zenith devices can run 30B+ parameter models locally, with additional security features for agentic applications. That includes OS-enforced identity, Microsoft Execution Containers (MXC), and enterprise-grade manageability for agents.
Frequently Asked Questions
Open a question for an answer from TweakTown's coverage of this news, or ask your own below.
Which Ryzen AI Halo systems will ship with Project Zenith preinstalled?
Which developer tools come preloaded in Project Zenith besides Visual Studio Code, WSL, GitHub Copilot CLI, Python 3.14+, Ubuntu, and PowerShell?
What are Microsoft Execution Containers and OS-enforced identity in Project Zenith?
How does Project Zenith help run 30B+ parameter models locally instead of in the cloud?
Have a question about this content not listed here? Ask below and TweakBot will answer it.
Beyond its developer-forward design, Project Zenith was created to address the current AI token issue, as costs have risen with more complex models and powerful agents being used more and more. With support for 30B+ parameter models, this helps shift some of that burden from the cloud to edge devices like Ryzen AI Halo systems running Project Zenith.
