Microsoft has introduced the Maia 300, a new multichiplet AI inference accelerator that will serve as the company's primary infrastructure for large-scale model processing starting next year.
Microsoft has announced the Maia 300, a 2nm chiplet-based AI accelerator set to launch in 2027 to power the company's internal AI infrastructure and reduce reliance on external GPU providers.
Microsoft has unveiled its third-generation AI chip, Maia 300, aiming to optimize cloud inference efficiency and support advanced reasoning models starting in 2027.
Microsoft has announced its third-generation custom AI chip, the Maia 300, which is set for production deployment in 2027 to power AI inference tasks and improve energy efficiency.
Microsoft is launching its third-generation Maia 300 AI chip, aiming to transition to in-house infrastructure for its cloud services and AI models by 2027.
