
NVIDIA, alongside Google and Microsoft, is advancing a new power architecture for AI compute called 800 VDC. This system simplifies power delivery by reducing conversion stages, allowing more efficient energy transfer from the grid to GPUs. The initiative is part of the Open Compute Project and involves over 80 companies developing compatible products. Existing AI facilities can adopt this architecture without major infrastructure changes, ensuring they remain competitive as compute demands grow. The move is seen as crucial for scaling AI infrastructure to meet future needs.
Read originalTopicNvidia Product And Business UpdatesCooling
Earlier coverage that leads up to this article, and what followed. Lines connect each piece to the closest one after it, converging here.
AI News · April 30, 2026 · Related
Matt Wolfe · May 11, 2026 · Related
The Rundown AI · June 2, 2026 · Related
MIT Technology Review AI · June 16, 2026 · Related
NVIDIA Blog · June 22, 2026 · Same story
The Verge AI · June 22, 2026 · Related
NVIDIA Blog · July 14, 2026 · Same story
WIRED AI · July 21, 2026 · Related
TechCrunch AI · August 13, 2026 · Related
WIRED AI · September 3, 2026 · Related
The Verge AI · September 3, 2026 · Related
Together AI Blog · September 10, 2026 · Related
TechCrunch AI · September 17, 2026 · Related