Los registros detrás de este evento
2-
Chip
Microsoft Maia 200
Microsoft's 3 nm inference accelerator: 216 GB of HBM3E and over 10 PFLOPS of FP4.
Maia 200, unveiled January 26, 2026, is Microsoft's second-generation in-house accelerator and is built for inference: more than 140 billion transistors on TSMC 3 nm, over 10 PFLOPS of FP4, 216 GB of HBM3E at 7 TB/s and 272 MB of on-chip SRAM in a 750 W package. It scales to 6,144 accelerators over an Ethernet-based two-tier network. Microsoft says it is in…
Abrir registro → -
Empresa
Microsoft
Azure, Copilot, a looser OpenAI alliance, and Fairwater-class AI superfactories.
Microsoft turned its OpenAI partnership into the fastest enterprise AI franchise: Azure hosts frontier training and serving, Copilot spreads models through Office and GitHub, and its Maia silicon and Fairwater AI data centers in Wisconsin and Atlanta push down the stack. An April 2026 amendment made its OpenAI IP license non-exclusive through 2032 and let…
Abrir registro →
