Nvidia Deepens Partnership With AI Chip Startup D-Matrix

Chip startup d-Matrix said Thursday it plans to use Nvidia’s networking hardware to connect its chips for running AI models to each other and to Nvidia’s Vera central processing units, which are used in data centers.

The startup will use Nvidia’s NVLink Fusion products, including switches, data processing units and network interface cards, to build server racks that contain both d-Matrix chips and Nvidia hardware. One server rack will hold 18 trays of d-Matrix chips and nine trays of Nvidia switches.

The company’s NVLink Fusion-enabled racks are set to ship in late 2027, according to d-Matrix CEO Sid Sheth. It’s unclear what the scale of the partnership will be.

The announcement shows how Nvidia is expanding sales to customers that don’t buy its flagship graphics processing units, including to competitors like d-Matrix. Switches and other networking gear are becoming an increasingly significant part of Nvidia’s revenue.

The d-Matrix partnership follows Amazon’s August announcement that it would be a larger customer of Nvidia’s NVLink Fusion products, and Nvidia’s $3.5 billion investment that same week in chip design firm MediaTek to expand NVLink Fusion production.

D-Matrix, founded in 2019, last raised $275 million at a $2 billion valuation in November and is in talks to fundraise again, The Information previously reported.