# Microsoft plans Maia 300 rollout for production AI workloads in 2027

_Published Monday, October 12, 2026 at 10:07 AM EDT · AI, Infrastructure · Latest · Tier 1 — Major_

![Maia 300 prototype chip 03 — Primary](https://platform.theverge.com/wp-content/uploads/sites/2/2026/10/Maia-300-prototype-chip-03.jpg?quality=90&strip=all&crop=0%2C10.732984293194%2C100%2C78.534031413613&w=1200)

Microsoft announced Maia 300, its third-generation in-house AI chip, with production deployment set to begin in 2027. The company plans to scale its Maia family to multi-gigawatt deployments starting that year and make it the primary system for running AI inference across Microsoft products and services.

Microsoft expects Maia 300 to deliver 40 percent more tokens per megawatt than leading GPU-based systems and up to 7x Maia 200's rack-scale token throughput in a 96-accelerator configuration. Built on TSMC's N2P process, its chiplet architecture separates compute, memory controllers and input/output functions. Those performance figures are Microsoft's expectations ahead of production rollout.

## Sources

- [The Verge](https://www.theverge.com/news/1009465/microsoft-maia-300-chip-improvements-rollout)

---
Canonical: https://techandbusiness.org/newswire/D9A_IVnG0USe02-tQjl0S_
Published: 2026-10-12T14:07:24.928Z
Story chronology: 2026-10-12T13:00:00.000Z
Retrieved: 2026-10-12T16:38:34.761Z
Publisher: Tech & Business (techandbusiness.org)