Microsoft는 차세대 Azure Maia AI 가속기 랙을 수 기가와트 규모로 도입할 계획이다. 96개 칩으로 구성된 이 제품은 전작 대비 토큰 처리량을 7배 높일 것으로 예상되며, 실제 운영 배포는 2027년으로 예정되어 있다.

Microsoft has revealed the next generation of its Azure Maia custom AI chips, saying they will usher in a new phase of compute infrastructure in which everything is closely integrated to meet the demands of agentic inference. The Redmond company says its Azure Maia 200, announced earlier this year, is now serving production workloads, and that a successor is already planned for 2027.

Azure Maia 300 is described as "a complete intelligent system" designed for the new types of AI workloads emerging today. Microsoft claims that a full 96-accelerator configuration will deliver up to seven times Maia 200's rack-scale token throughput, while offering 40 percent more tokens per megawatt than leading GPU-based systems. The next-gen chip has an inference-first architecture optimized for long-context, memory-intensive and decode-heavy AI workloads, according to a blog post by Scott Guthrie, Microsoft's EVP for Cloud and AI.

The company envisions deploying the chip in rack-scale systems of up to 96 accelerators. Each Maia 300 will feature eight HBM4 memory stacks providing up to 18 TB/s of peak memory bandwidth, to support the execution of large models and long-context workloads.

This resembles the rack-scale AI equipment already developed by Nvidia for its GPU accelerators, and more recently by AMD with its Helion and Helios infrastructure, all designed to make the most of their own specific silicon and software stacks. Microsoft says Maia 300 is "built as a complete system, with software, silicon, physical rack infrastructure, networking, and datacenters co-designed to operate as a single platform to meet the demands of cloud-scale agentic inference." The company argues that this system-level approach is necessary for agentic workloads, which involve multi-step execution, long-context decoding and tool use, and that they "change the computational profile and arithmetic intensity of inference."

The growing emphasis on in-house integration contrasts with Microsoft's 2024 "AI Access Principles," which stressed partnerships across its technology stack. The company even complained to the European Commission's competition authorities that Google had a competitive advantage because of its approach. "Only one company – Google," Microsoft said at the time, "is vertically integrated in a manner that provides it with strength and independence at every AI layer from chips to a thriving mobile app store. Everybody else must rely on partnerships to innovate and compete."

Microsoft now has its own AI chips, rack-scale designs and internally developed MAI models, giving it more control over its stack. It plans to scale Azure Maia to "multi-gigawatt deployments" starting in 2027.

Microsoft was asked for comment. Guthrie said the company will share more details on its strategy at next month's Microsoft Ignite 2026 event in San Francisco.