d-Matrix Adopts NVIDIA NVLink Fusion for Rack-Scale XPU Deployment
d-Matrix, an AI inference chipmaker, announced it will use NVLink Fusion to connect its next-generation Raptor XPUs to NVIDIA’s AI infrastructure platform, enabling faster and lower-risk deployment.
AI inference chipmaker d-Matrix announced it will use NVLink Fusion to connect its next-generation Raptor XPUs to NVIDIA’s AI infrastructure platform, joining a growing roster of ecosystem partners. By integrating Raptor XPUs with NVIDIA’s MGX rack architecture and the broader NVIDIA AI platform, d-Matrix can accelerate time to market and provide a lower-risk path for large-scale deployment.
"Demand for inference is soaring, but capital, time and energy remain finite," said Sid Sheth, cofounder and CEO of d-Matrix during a press briefing. "With NVLink Fusion and MGX, we can integrate our Raptor XPUs into a broadly deployed, liquid-cooled architecture, giving customers a faster, lower-risk path to deploy and scale ultralow-latency inference."
NVLink Fusion is a high-bandwidth, low-latency technology that connects custom XPUs and CPUs to NVIDIA’s infrastructure. It enables d-Matrix to tap into the NVIDIA MGX ecosystem’s mature, validated rack designs, supply chain, power, and cooling infrastructure. This standardization allows data centers to support GPUs, CPUs, and XPUs within a single rack architecture.
Using NVIDIA NVLink, d-Matrix plans to connect its XPUs in a single high-bandwidth, low-latency scale-up domain. Its racks can also work alongside NVIDIA GPU-based systems like NVIDIA Vera Rubin NVL72 for disaggregated inference. Additionally, d-Matrix plans to integrate NVIDIA Vera CPUs, NVIDIA ConnectX-9 SuperNICs, NVIDIA BlueField-4 DPUs, and NVIDIA Spectrum-X Ethernet networking.
NVLink Fusion provides a proven foundation for deploying specialized inference alongside NVIDIA systems within flexible, unified AI factories. This technology opens access to NVIDIA’s full-stack AI factory platform, designed to run every AI workload with the best performance per watt and lowest cost per token. Silicon innovators like d-Matrix can use NVLink Fusion to increase performance, accelerate time to market, and reduce the risk of deploying semi-custom AI factories.
Source: nvidia-newsroom
