Microsoft’s new Maia 300 packs 96 chips into one 167-kW rack with 40% token efficiency

What happened

Microsoft expects GitHub Copilot and Microsoft Foundry to adopt the hardware by the end of 2026. Microsoft plans to break free from NVIDIA and deploy its next-generation Maia 300 AI accelerator in 2027, targeting 40% more tokens per megawatt than leading GPU (a chip built for many calculations at once, used for graphics and AI)-based systems. Microsoft is a software and software development company based in Redmond.

The chip will power Microsoft’s expanding AI infrastructure, with a 96-accelerator rack configuration designed to deliver up to seven times the token throughput (how much work a system gets through in a given time) of its predecessor. The announcement marks another step in Microsoft’s effort to build custom silicon for large-scale AI inference.

Its Maia 200 accelerator already handles production workloads, including Microsoft Copilot. Maia 300 takes a system-level approach to inference efficiency.