Torsten Hoefler on Maia 300: ‘Our SDLA is a great architecture for LLMs’
Torsten Hoefler 🇨🇭 @thoeflerX
Why it matters
First-hand comment from an ETH Zurich professor who links the Maia architecture paper (arXiv 2608.24664) and hints at further specifications.
Summary
Posted Oct 12, 2026 (14:48 UTC), quoting @satyanadella (https://x.com/satyanadella/status/2109631106404487523). Hoefler (ETH Zurich, head of SPCL; his profile does not state a Microsoft role) quotes Nadella’s announcement: ‘Maia is going strong!! Next chip coming online soon at multi-gigawatt deployment. Wait for the next record breaking specs. Our SDLA is a great architecture for LLMs’, with a link to arXiv 2608.24664. The paper is “Maia 200: A Software Defined Dataflow System for Large-scale AI Acceleration” (posted 25 Aug 2026; title and date from the arXiv API, body not read).
Metrics as of 2026-10-13 via api.fxtwitter.com: 1,525 views, 27 likes, 0 reposts. Author: about 6,227 followers.
Archived text
Maia is going strong!! Next chip coming online soon at multi-gigawatt deployment. Wait for the next record breaking specs 😉.
Our SDLA is a great architecture for LLMs 😀.
https://arxiv.org/abs/2608.24664
Quoting @satyanadella: Today we are announcing our biggest bet yet on our own silicon.
Starting next calendar year, we will scale our Maia family to multi-gigawatt deployments, making it the primary inference system for our own product and services.
And we’re introducing Maia 300, which delivers 40% more tokens per watt compared to leading GPU-based systems.
We are already deploying Maia 200 across the US, Europe, and Australia, delivering up to 1.7x more tokens per watt across experiences in Teams and Copilot.
Our goal is to build on the full-stack efficiency gains we’ve already seen, while continuing to bring the latest innovations from across the ecosystem into our fleet.
Cited in
People in this post
Satya Nadella, Chairman and CEO, Microsoft