Microsoft introduced Maia 300 on October 12, 2026, its third-generation custom AI accelerator for inference. The company says production deployments will begin in 2027 and that Maia will become the primary inference system for Microsoft products and services as deployments expand to multi-gigawatt scale.

According to Microsoft’s technical disclosure, each Maia 300 uses eight HBM4 stacks delivering 18.0 TB/s of peak memory bandwidth within a 950-watt TDP. A full rack is designed around 96 accelerators, 27.6 TB of HBM and roughly 167 kW of power.

Microsoft says Maia 300 can deliver 40% more tokens per megawatt than “leading GPU-based systems” and up to seven times the rack-scale token throughput of Maia 200. Those are vendor claims; Microsoft has not identified the specific GPU baseline for the 40% comparison, so independent testing will be needed before treating the efficiency figure as a general performance result.

Reuters reported that Microsoft may deploy more than one million Maia chips in 2027, underscoring the scale of the internal-silicon strategy.

The announcement fits a broader shift toward rack-level AI design. BitcoinVersus.Tech has recently covered Intel’s high-memory GPU D-Series, Astera Labs’ modular Orion rack and Delta’s 800 VDC power and megawatt-scale cooling work.

Today we are announcing our biggest bet yet on our own silicon.

Starting next calendar year, we will scale our Maia family to multi-gigawatt deployments, making it the primary inference system for our own product and services.

And we’re introducing Maia 300, which delivers 40%… pic.twitter.com/L490XvOys9

— Satya Nadella (@satyanadella) October 12, 2026

BitcoinVersus.Tech Editor’s Note:
We volunteer daily to ensure the credibility of the information on this platform is Verifiably True. If you would like to support to help further secure the integrity of our research initiatives, please donate using the QR code provided