NVIDIA· Updated
NVIDIA and Marvell Technology partnered to integrate custom silicon into the NVIDIA AI factory ecosystem through the NVLink Fusion platform. This allows customers to build semi-custom AI infrastructure that remains fully compatible with NVIDIA's proprietary interconnects and software stack.
NVLink interconnects alongside NVIDIA's own hardware.As enterprises move toward specialized "AI factories," many develop custom silicon to optimize specific workloads. Previously, using non-NVIDIA chips often meant losing access to the dominant NVLink fabric. This partnership bridges that gap, allowing specialized hardware to leverage NVIDIA’s global supply chain and software stack without ecosystem fragmentation.
You can now leverage Marvell’s leadership in optical interconnects and silicon photonics within standard NVIDIA architectures. The collaboration also targets the telecommunications sector through NVIDIA Aerial AI-RAN, aiming to convert 5G and 6G networks into distributed AI infrastructure. These solutions are designed for hyperscalers building next-generation scale-up networking.
NVIDIA and @MarvellTech join forces through NVLink Fusion to expand ecosystem and give customers greater choice and flexibility in developing next-generation infrastructure. https://t.co/peHkhGqcKS
Every HeadsUpAI update is written based on its original source and reviewed before it's published. Read our editorial standards →
