Newest
NVIDIA claims Vera Rubin NVL72 delivers over 30x throughput per megawatt versus GB300 NVL72 for inference
Running large AI models is suddenly far cheaper and far more power-efficient for data centers.
Score 9.2,
AI Infrastructure
More stories
Newest first.
5 of 9 stories
d-Matrix adopts NVIDIA NVLink Fusion to link Raptor inference accelerators into NVIDIA AI racks
Score 9.3,
AI Infrastructure
Oracle Q1 FY27 results highlight deployment of over 300,000 GPUs and 850 MW of AI capacity, with agentic AI accelerator planned
Score 9.3,
AI Infrastructure
AMD unveils Threadripper Halo Station, a deskside AI ‘personal supercomputer’ with Instinct MI350P accelerators
Score 9.2,
AI Infrastructure
NVIDIA invests $3.5B in MediaTek and opens NVLink Fusion to custom AI accelerators across rack-scale infrastructure
Score 9.6,
AI Infrastructure