Microsoft to expand Azure Maia system to multi-gigawatt deployments
Azure Maia 200 is now in production, and the Azure Maia 300 chip is scheduled for release in 2027. Microsoft plans to scale these systems to multi-gigawatt deployments to serve as the primary inference infrastructure for its services.
Verified State Diff
Impact & Verification Analysis
Azure enterprise customers, AI developers, and cloud infrastructure architects.
This represents a major strategic move to reduce dependency on external GPU suppliers (like NVIDIA) and optimize cost-to-performance ratios for large-scale AI inference workloads.
Full Fact Overview
Microsoft is vertically integrating its AI infrastructure by transitioning from reliance on third-party silicon to its proprietary Maia custom-silicon architecture. The Maia 200, an AI accelerator designed for large language model training and inference, has reached production status. The roadmap confirms the development of the Maia 300, which will be deployed at a multi-gigawatt scale by 2027, signaling a strategic shift to prioritize internal hardware for Microsoft's AI service stack over general-purpose cloud hardware.