NuAxiSemiconductor trade intelligence
Microsoft's Maia 300 rack: 96 accelerators, HBM4, mostly liquid-cooled
What happened
Microsoft has released architecture details for Maia 300, its third-generation Azure AI accelerator: the design pairs 96 accelerators per rack, uses HBM4, and is mostly liquid-cooled. A two-tier network links up to 12,288 accelerators into a single superpod. The company says the system is built for cloud-scale agentic inference and that it co-designed the software, silicon, rack infrastructure, networking, and data centers to run as one platform.
Impact and signal
Low impactImpact score: 10/100
Affected components and suppliers
Maia 300 acceleratorHBM4
Maia 300Azure AI acceleratorrack architectureHBM4liquid coolingagentic inference
Source and time
Source: DigiTimes Asia · 10/12/2026 · View original source
Open the NuAxi live feed →
Real-time pricing, supply signals and AI analysis for semiconductor traders