d-Matrix adopts NVIDIA NVLink Fusion for rack-scale Raptor XPU inference

BenBajarin · x · 2026-09-11

Inference chipmaker d-Matrix will connect its next-gen Raptor XPUs to NVIDIA's AI platform via NVLink Fusion, combining NVLink scale-up, Spectrum-X scale-out and MGX rack architecture for lower-risk, ultralow-latency inference deployment. Analyst Ben Bajarin says this confirms ASICs as complementary to NVIDIA's compute fabric, not islands.

Related event: d-Matrix taps NVIDIA NVLink Fusion for rack-scale Raptor XPU inference(2 posts)→

Original post →

More from Infra

Infra channel →