d-Matrix Adopts NVIDIA NVLink Fusion for Rack-Scale XPU Deployment

d-Matrix, an AI inference chipmaker, has announced its adoption of NVIDIA's NVLink Fusion for deploying its Raptor XPUs in a rack-scale XPU architecture. This move places d-Matrix alongside a growing list of ecosystem partners, including AWS, Arm, Intel, Fujitsu, and more. By integrating its Raptor XPUs with NVIDIA's NVLink scale-up networking, Spectrum-X scale-out networking, MGX rack architecture, and the broader NVIDIA AI platform, d-Matrix will benefit from an accelerated, lower-risk path to large-scale deployment.

Sid Sheth, co-founder and CEO of d-Matrix, highlighted the soaring demand for inference and the limited resources of capital, time, and energy. With NVLink Fusion and MGX, d-Matrix can integrate its Raptor XPUs into a widely deployed, liquid-cooled architecture, providing customers with a faster, lower-risk route to deploy and scale ultralow-latency inference.

NVLink Fusion connects third-party custom XPUs and CPUs to NVIDIA's AI infrastructure platform, allowing silicon companies to concentrate on processor innovations while leveraging NVIDIA's infrastructure for deployment at AI factory scale. This platform is both vertically integrated and horizontally open, offering flexibility in matching the right compute to each workload within a common AI factory.