Microsoft Says Maia 300 Delivers 40 Percent More Tokens per Megawatt in 2027
Summary
Microsoft says Azure Maia 200 now serves Copilot production workloads and that it will scale the Maia family to multi-gigawatt deployments starting in 2027. Scott Guthrie claims Maia 300, due in production that year, delivers 40 percent more tokens per megawatt than leading GPU systems across 96 accelerators per rack.
Key Points
- Microsoft says Maia 200 deploys across the United States, Europe and Australia, delivering up to 1.7x more tokens per watt than GPU infrastructure.
- Scott Guthrie says Maia 300 delivers up to 7x the rack-scale token throughput of Maia 200.
- Maia 300 uses eight HBM4 stacks delivering 18.0 TB/s peak memory bandwidth within a 950-watt TDP.