# d-Matrix Will Plug Its Inference Chips Into NVIDIA Racks Through NVLink Fusion

Up to 144 Raptor accelerators can share a single NVLink domain, with rack-integrated systems expected in the fourth quarter of 2027.

- Published: 2026-09-11T06:22:22.078Z
- Canonical: https://polylog.news/ai/2026-09-11/d-matrix-will-plug-its-inference-chips-into-nvidia-racks-thr
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [NVIDIA Blog](https://blogs.nvidia.com/blog/d-matrix-nvlink-fusion/), [StorageReview](https://www.storagereview.com/news/d-matrix-raptor-xpus-join-nvidia-mgx-racks-through-nvlink-fusion-first-systems-due-q4-2027), [PR Newswire](https://www.prnewswire.com/news-releases/d-matrix-adopts-nvidia-nvlink-fusion-rackscale-infrastructure-for-ultra-low-latency-ai-inference-302875104.html)

d-Matrix, which builds accelerators specialized for inference rather than training, said it will connect its next-generation Raptor processing units to NVIDIA's infrastructure using NVLink Fusion, NVIDIA's program for letting third-party si…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-09-11/d-matrix-will-plug-its-inference-chips-into-nvidia-racks-thr (subscription information: https://polylog.news/pricing).