Microsoft says its third-generation custom AI chip will become the primary inference system for its own products and services. Maia 200 is now in production.
Microsoft says it will scale its Azure Maia chip family to multi-gigawatt deployments starting in 2027, making Maia the primary inference system for Microsoft products and services. Maia 300, the third generation, is built for long-context, memory-heavy and agentic workloads, and begins production deployments in 2027. Microsoft says a 96-accelerator rack runs in a 167 kW power envelope with 27.6 TB of HBM and 1.7 PB/s of aggregate HBM bandwidth. Microsoft expects 40% more tokens per megawatt than leading GPU-based systems and up to 7x the rack-scale token throughput of Maia 200, which it says is now in production. These are company projections, not independent benchmarks. Microsoft has said it will share more at Ignite next month.
Sources: Microsoft Blog · Azure Blog