D-Matrix, a US-based AI inference chipmaker, said on September 10 it will use Nvidia NVLink Fusion to connect its next-generation Raptor XPUs to Nvidia’s AI infrastructure platform for rack-scale deployments.

The company said it plans to use Nvidia NVLink scale-up, Spectrum-X scale-out networking.

It will also the Nvidia MGX rack architecture so Raptor can run in larger AI clusters and with Nvidia graphics processing unit-based systems.

Nvidia has described NVLink Fusion as a way to build semi-custom AI infrastructure with partners, giving third-party processors a path into Nvidia-based racks and systems.

D-Matrix has raised US$275 million. The company has also said its Corsair AI Inference Platform has entered full production to meet customer demand.

🔗 Source: Nvidia

Recent d-Matrix developments