AI

Microsoft announces Maia 300 AI chip and multi-gigawatt Azure deployment plans

Microsoft announced Maia 300, its next-generation custom AI inference chip, and plans to begin multi-gigawatt Azure deployments in 2027. Microsoft says a rack configuration with 96 chips is planned; one account says it will use about 167 kW of power, while another says the configuration promises seven times the token throughput of its predecessor. The company plans to make Maia 300 its primary AI inference system for Microsoft products and services, with production workloads scheduled to begin in 2027.

AI-written summary, checked against the cited sources. How summaries work

Image: HPCwire

Why it matters

Microsoft says Maia 300 is intended to become the primary AI inference system for its products and services, with production deployments starting in 2027.

What we know

Confirmed by several sources

Reported by one source

  • A 96-chip rack configuration is planned, and one report says it promises seven times its predecessor's token throughput. — The Register
  • A 96-chip rack will use about 167 kW of power. — Techmeme
  • Maia 300 is built on TSMC's N2P process node. — Techmeme
  • Microsoft plans to make Maia 300 its primary AI inference system for its products and services. — HPCwire

Coverage 3 publishers

  1. TechmemeFirst report

    Microsoft unveils its third-gen in-house AI chip, the Maia 300, built on TSMC's N2P process node and aimed at serving as its “primary AI inference system”

    The new Maia 300 chip will start rolling out to production AI workloads in 2027.

  2. HPCwire

    Microsoft Announces Maia 300, Its Next-Gen Custom AI Chip

    Oct. 12, 2026 — Microsoft has announced Maia 300, the next generation of its AI inference platform, and plans to expand Azure Maia to multi-gigawatt deployments beginning in 2027. With the rollout, Maia 300 will become the primary AI…

  3. The Register

    Microsoft plans multi-gigawatt rollout of next-gen Azure Maia AI accelerator racks

    96-chip configuration promises seven times its predecessor's token throughput, with production deployments due in 2027

Earliest report first. Articles stay on their publishers’ sites; each link opens the original.