ChipsSeptember 10, 2026via NVIDIA Blog
d-Matrix Adopts NVIDIA NVLink Fusion for Rack-Scale XPU Deployment
Why it matters
d-Matrix's adoption of NVIDIA NVLink Fusion signals continued ecosystem consolidation around NVIDIA's interconnect standard for AI inference at scale. It reflects the industry's bet on standardized, vendor-agnostic rack architecture—but also deepens lock-in.
Key signals
- d-Matrix Raptor XPUs now integrate NVIDIA NVLink Fusion for scale-up
- Integration includes Spectrum-X scale-out networking
- Uses NVIDIA MGX rack architecture
- Part of broader NVIDIA AI platform ecosystem
- Announced Sept 10, 2026
- d-Matrix is an AI inference chipmaker, competing in post-Nvidia specialization space
- d-Matrix Raptor XPU will use NVIDIA NVLink Fusion for interconnect
- Integration with NVIDIA MGX rack architecture and broader NVIDIA AI platform
- NVLink Fusion combines Spectrum-X scale-out networking for disaggregated inference
- Part of growing ecosystem of non-NVIDIA chip makers adopting NVIDIA interconnect standards
- Announcement published on NVIDIA blog (first-party source)
The hook
d-Matrix locks into NVIDIA's ecosystem standard for next-gen inference hardware—another bet on NVLink dominance in the rack.
AI inference chipmaker d-Matrix today announced it will use NVIDIA NVLink Fusion to connect its next-generation Raptor XPUs to NVIDIA’s AI infrastructure platform — joining a growing roster of ecosystem partners. By connecting Raptor to NVIDIA NVLink scale-up and Spectrum-X scale-out networking, the…