d-Matrix will use Nvidia’s chip-linking technology in AI inference servers

The startup plans to connect its Raptor inference chips into Nvidia data-center systems using NVLink Fusion.

AI chip startup d-Matrix says its next-generation inference processors will connect directly into Nvidia server systems through NVLink Fusion. The partnership reflects growing demand for specialized chips optimized for running AI models at low latency.

Why it matters

This development matters to builders, founders and technology teams because it changes the competitive, technical or economic landscape around modern software and AI.

What to watch

CodaDaily will continue tracking product releases, adoption, funding, regulation and the practical impact on developers and businesses.

Source

This CodaDaily article is an original summary based on the source linked below; it does not reproduce the source article.