d-Matrix adopts NVIDIA NVLink Fusion for XPU racks

Inference chipmaker d-Matrix will use NVIDIA NVLink Fusion to connect its Raptor XPUs to NVIDIA’s AI platform and MGX rack architecture, easing rack-scale deployment. NVLink Fusion claims 3x lower XPU-to-XPU latency versus Ethernet, 10x higher packet rates, and 3 TB/s per XPU all-to-all bandwidth.

• Nvidia partner GMI Cloud seeks $300M loan for chips2026-09-17
• NVIDIA advances native GPU programming in Rust2026-09-17
• Huawei to unveil China’s answer to Nvidia AI chips2026-09-17
• Nvidia: robots await a ChatGPT moment2026-09-16
• NVIDIA Vera Rubin NVL72 tops MLPerf Inference v6.12026-09-16