Podcast
Oct 13, 2026 · 6 min · 7 segments
Microsoft says its next AI chip system will produce 40% more tokens per megawatt. Production deployments are due to start next year. This edition covers 5 evidence-backed stories. AI-generated…
Microsoft announced Maya 300 on October 12 and plans production deployments in 2027.
It expects the system to deliver 40% more tokens per megawatt than leading GPU-based systems.
Microsoft also expects up to seven times Maya 200's token throughput in a full rack.
A Maya 300 rack combines 96 accelerators within a 167 kilowatt power envelope.
Maya 300's design separates computing, memory control, and input-output work into different chiplets.
Those figures measure different things.
One measures power efficiency, while the other measures total rack output.
Microsoft has not published independent results or promised lower customer prices.
Maya 300 is due in 2027, while Maya 200 runs production workloads today.
Microsoft's 2027 production results need to show which models run on Maya 300 and how much energy each useful answer takes.
A rollout date answers neither question.
Microsoft announced Maya 300 on October 12 and plans production deployments in 2027.
It expects the system to deliver 40% more tokens per megawatt than leading GPU-based systems.
Microsoft also expects up to seven times Maya 200's token throughput in a full rack.
A Maya 300 rack combines 96 accelerators within a 167 kilowatt power envelope.
Maya 300's design separates computing, memory control, and input-output work into different chiplets.
Those figures measure different things.
One measures power efficiency, while the other measures total rack output.
Microsoft has not published independent results or promised lower customer prices.
Maya 300 is due in 2027, while Maya 200 runs production workloads today.
Microsoft's 2027 production results need to show which models run on Maya 300 and how much energy each useful answer takes.
A rollout date answers neither question.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.
