Microsoft plans to deploy racks of its next-generation Azure Maia AI accelerators in a multi-gigawatt rollout, with production deployments due in 2027.
Microsoft plans to make its in-house Azure Maia chips its main AI inference system across its products starting in 2027 at multi-gigawatt scale, with the upcoming Maia 300 said to deliver 40 percent more tokens per megawatt than leading GPU systems.
Why it matters: It signals Microsoft intends to shift the bulk of its AI inference off third-party GPU systems and onto its own silicon.
Written by AI from The Decoder's published text. Read the original for full details.
More coverage of this story (2)
Microsoft plans multi-gigawatt rollout of next-gen Azure Maia AI accelerator racks
Primary source
Reporting from a publisher outside the organisation being covered.
Read original story ↗