D-Matrix adopts Nvidia NVLink Fusion for Raptor chips
D-Matrix said it will adopt Nvidia's NVLink Fusion chip-linking technology so its Raptor inference chips can plug directly into Nvidia data-center racks, aiming to deliver fast, low-latency AI services such as coding assistants, chatbots and voice agents. Raptor chips are scheduled to finish final design by the end of this year, and Nvidia-compatible server racks are expected to be available in 2027. The startup also partners with Astera Labs for connectivity, has Microsoft backing from prior financing, and did not disclose the financial terms of the collaboration.
Raptor chips will plug into Nvidia racks using NVLink Fusion.
Context
AI workloads are shifting from model training to model inference. D-Matrix shipped its first AI chip in November 2024 and will finish Raptor design by year end. Nvidia-compatible racks are expected in 2027, so product availability may…
The full analysis
19 dimensions on this story — world impact, market read, and what happens next.
- Full ContextLocked
- Affected SectorsLocked
- Stock ImpactLocked
- Economic IndicatorLocked
- Investor RelevanceLocked
- Professional RelevanceLocked
- Watch PointsLocked
- Probability of ChangeLocked
- Debate PointsLocked
- Prerequisite KnowledgeLocked
- Follow-up QuestionsLocked
- Pros & ConsLocked