Microsoft plans to deploy its Azure Maia AI chips at a multi-gigawatt scale starting in 2027, making them the company's primary inference system across its products. Executive Scott Guthrie confirmed the rollout, noting that the current Maia 200 chip already powers Microsoft Copilot in the US, Europe, and Australia, with GitHub Copilot and Foundry integrations scheduled to follow by the end of the year. CEO Satya Nadella described the initiative as the company's biggest bet yet on its own custom silicon.
The architecture is designed to handle OpenAI models, MAI models, and open-source models, delivering up to 1.7 times more tokens per watt than leading GPUs in Microsoft Teams workloads. Looking ahead, the successor Maia 300 chip is slated to ship in 2027, specifically targeting long inputs, fast chats, and agentic tasks. Microsoft claims the Maia 300 will deliver 40 percent more tokens per megawatt than leading GPU systems, with a 96-chip rack producing up to seven times the token output of a Maia 200 rack.
This hardware scaling initiative highlights Microsoft's strategy to reduce reliance on merchant GPUs and optimize efficiency for large-scale AI inference workloads.