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

AI inference chipmaker d-Matrix has revealed plans to utilize NVIDIA NVLink Fusion technology to connect its upcoming Raptor XPUs to NVIDIA’s comprehensive AI infrastructure platform. This integration aims to enable high-performance, scalable deployment of XPUs at rack scale, leveraging NVIDIA’s MGX rack architecture and Spectrum-X networking solutions.

By adopting NVLink Fusion, d-Matrix joins a growing ecosystem of partners that are enhancing AI infrastructure capabilities through NVIDIA’s scalable interconnect technology. The integration is designed to facilitate efficient data transfer and communication between XPUs, which is critical for large-scale AI inference workloads.

The Raptor XPUs are part of d-Matrix’s next-generation offerings, and their connection via NVLink Fusion is expected to improve throughput and reduce latency in AI inference tasks. This development aligns with NVIDIA’s broader strategy to enable flexible, high-performance AI infrastructure solutions for enterprise and data center deployments.

While specific deployment timelines and performance benchmarks are not disclosed, the collaboration underscores the importance of advanced interconnect technologies in supporting the growing demands of AI workloads. The integration also demonstrates NVIDIA’s ongoing efforts to expand its AI ecosystem through partnerships that leverage its infrastructure architecture.