d-Matrix to Connect Raptor XPUs to NVIDIA Platform Using NVLink Fusion
d-Matrix announced it will use NVIDIA NVLink Fusion to connect its next-generation Raptor XPUs to NVIDIA's AI infrastructure platform.
Quick answer
What did d-Matrix announce about NVIDIA NVLink Fusion?
d-Matrix announced it will use NVIDIA NVLink Fusion to connect its next-generation Raptor XPUs to NVIDIA's AI infrastructure platform, including NVLink scale-up and Spectrum-X networking and the MGX rack architecture. The company said this gives it a path to deploy Raptor at large scale alongside NVIDIA systems.
Key takeaways
- d-Matrix announced it will connect its next-generation Raptor XPUs to NVIDIA's AI infrastructure platform using NVLink Fusion.
- The integration links Raptor to NVIDIA NVLink scale-up networking, Spectrum-X scale-out networking, and the NVIDIA MGX rack architecture.
- d-Matrix said its racks can operate alongside NVIDIA GPU-based systems such as NVIDIA Vera Rubin NVL72 for disaggregated inference.
- d-Matrix said it also plans to integrate NVIDIA Vera CPUs, ConnectX-9 SuperNICs, BlueField-4 DPUs, and Spectrum-X Ethernet networking.
- d-Matrix CEO Sid Sheth said the approach gives customers a faster, lower-risk path to deploy and scale ultralow-latency inference.
AI inference chipmaker d-Matrix announced it will use NVIDIA NVLink Fusion to connect its next-generation Raptor XPUs to NVIDIA's AI infrastructure platform, according to an NVIDIA blog post.
The company said the connection links Raptor to NVIDIA NVLink scale-up networking, Spectrum-X scale-out networking, the NVIDIA MGX rack architecture, and the broader NVIDIA AI platform.
What d-Matrix Said
"Demand for inference is soaring, but capital, time and energy remain finite," said Sid Sheth, cofounder and CEO of d-Matrix, during a press briefing according to NVIDIA. "With NVLink Fusion and MGX, we can integrate our Raptor XPUs into a broadly deployed, liquid-cooled architecture, giving customers a faster, lower-risk path to deploy and scale ultralow-latency inference."
NVIDIA said NVLink Fusion is the high-bandwidth, low-latency technology that connects custom XPUs and CPUs to the NVIDIA stack, and that it extends the openness of NVIDIA's AI platform to third-party silicon.
Planned Integrations
According to NVIDIA, d-Matrix plans to:
- Connect its XPUs in a single high-bandwidth, low-latency scale-up domain using NVIDIA NVLink
- Operate its racks alongside NVIDIA GPU-based systems such as NVIDIA Vera Rubin NVL72 for disaggregated inference
- Integrate NVIDIA Vera CPUs, ConnectX-9 SuperNICs, BlueField-4 DPUs, and Spectrum-X Ethernet networking
NVIDIA said these components, combined with NVLink and MGX, give d-Matrix a foundation for deploying specialized inference hardware within what NVIDIA calls unified AI factories.
Platform Context
NVIDIA described its full-stack AI factory platform as including NVIDIA Vera Rubin NVL72, Groq 3 LPX, the Vera CPU rack, Vera BlueField-4 STX storage, and Spectrum-6 SPX Ethernet networking, saying it is designed to run AI workloads and model architectures with the company's targeted performance-per-watt and cost-per-token goals. NVIDIA said NVLink Fusion gives customers the flexibility to match compute to workloads within that common platform.
Source: NVIDIA Blog, "d-Matrix Adopts NVIDIA NVLink Fusion for Rack-Scale XPU Deployment," published September 10, 2026.
Frequently asked questions
- What is d-Matrix connecting to NVIDIA's platform?
- d-Matrix is connecting its next-generation Raptor XPUs to NVIDIA's AI infrastructure platform using NVLink Fusion.
- What NVIDIA technologies does the integration involve?
- d-Matrix said it plans to use NVLink scale-up networking, Spectrum-X scale-out networking, the MGX rack architecture, Vera CPUs, ConnectX-9 SuperNICs, BlueField-4 DPUs, and Spectrum-X Ethernet.
- What did d-Matrix's CEO say about the move?
- Sid Sheth, cofounder and CEO of d-Matrix, said NVLink Fusion and MGX let the company integrate Raptor XPUs into a broadly deployed, liquid-cooled architecture, giving customers a faster, lower-risk path to deploy and scale ultralow-latency inference.
- Can d-Matrix's racks work with NVIDIA GPU systems?
- d-Matrix said its racks can work alongside NVIDIA GPU-based systems like NVIDIA Vera Rubin NVL72 for disaggregated inference.