d-Matrix and NVIDIA Partner to Integrate Raptor XPUs with NVIDIA AI Infrastructure for Enhanced Inference
Covered by 2 sources
d-Matrix and NVIDIA announced a multi-year collaboration to integrate d-Matrix's Raptor inference XPUs into NVIDIA's MGX AI infrastructure using NVLink Fusion, targeting late 2027 availability. This partnership aims to deliver a disaggregated AI inference solution combining SRAM-based chips and NVIDIA GPUs to improve inference speed and provide low-latency services to AI labs and hyperscalers.