Chip Startup D-Matrix to Use Nvidia Chip-Linking Tech in AI Servers
The startup expects systems with up to 144 Raptor accelerators linked by a single NVLink fabric by the end of next year.
- On Thursday, AI infrastructure startup d-Matrix announced it will license Nvidia's NVLink Fusion interconnect technology to integrate its Raptor accelerators directly into Nvidia's rack-scale reference designs.
- By adopting Nvidia's proven MGX ecosystem, d-Matrix addresses rising inference demand and sidesteps scaling challenges associated with its unique memory-centric chip architecture across large compute clusters.
- Raptor accelerators utilize a first-of-its-kind 3D DRAM stacking approach to boost memory bandwidth, with connectivity firm Astera Labs providing custom solutions to ensure seamless data flow within the ecosystem.
- Systems integrated with Raptor chips are expected to be available in Q4 2027, targeting latency-sensitive applications like AI coding assistants and real-time chatbots where speed is critical.
- Nvidia CEO Jensen Huang stated that NVLink Fusion "enables partners to integrate custom silicon" into infrastructure, a strategy expanding accelerator choices for AI factories while maintaining Nvidia's central ecosystem role.
14 Articles
14 Articles
The d-Matrix startup announced on Thursday (10) that it will adopt Nvidia chip interconnection technology to use its processors directly in the data center systems of the technology giant, as the demand for AI intensifys. Artificial intelligence workloads have been shifting from AI model training to the execution of these models in everyday use, a process known as inference. While Nvidia's expensive graphic processors dominate the training workl…
d-Matrix Joins Nvidia’s AI Inference Push
Nvidia is embracing the rise of custom AI chips as demand shifts toward specialized systems built for inference. d-Matrix CEO Sid Sheth discusses the startup’s new partnership with Nvidia, which will integrate its inference-focused XPUs with Nvidia GPUs using NVLink Fusion, and why inference is emerging as the next major battleground in AI compute. He joins Ed Ludlow on "Bloomberg Tech." (Source: Bloomberg)
Chip startup d-Matrix to use Nvidia chip-linking tech in AI servers
D-Matrix said on Thursday it will adopt Nvidia's chip-linking technology to use the startup's processors directly inside the semiconductor giant's data-center systems as demand for AI intensifies.
Nvidia Stock Price Falls 2.3% Despite d-Matrix Partnership
The post Nvidia Stock Price Falls 2.3% Despite d-Matrix Partnership appeared first on Coinpedia Fintech News Nvidia stock fell 2.35% Thursday even after announcing a collaboration with AI inference chipmaker d-Matrix. Under the deal, d-Matrix will use Nvidia’s NVLink Fusion to connect its next-generation Raptor XPUs to Nvidia’s AI platform, giving the chipmaker a path to deploy its custom chips at large AI data centers. Raptor will rely on Nvidi…
Coverage Details
Bias Distribution
- 86% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
