NVIDIA AI Ecosystem Expands as Marvell Joins Forces Through NVLink Fusion

NVIDIA's Endeavor building in Santa Clara, California

NVIDIA and Marvell Technology, Inc. (NASDAQ: MRVL) today announced a strategic partnership to connect Marvell to NVIDIA's AI factories and AI-RAN ecosystem through NVIDIA NVLink Fusion™, providing customers building on NVIDIA architectures with more choice and flexibility in developing next-generation infrastructure. The companies will also collaborate on silicon photonics technology.

In addition, NVIDIA will invest $2 billion in Marvell.

The foundation of the partnership is NVIDIA NVLink Fusion, a rack-scale platform that can develop semi-custom AI infrastructure leveraging the NVIDIA NVLink™ ecosystem. Marvell will provide custom XPU and NVLink Fusion-compatible scale-up networking, and NVIDIA will provide supporting technologies such as Vera CPU, ConnectX® NIC, BlueField® DPU, NVLink interconnect, Spectrum-X™ switch, and rack-scale AI computing.

For customers developing custom XPUs, NVLink Fusion enables heterogeneous AI infrastructure that is fully compatible with NVIDIA systems, allowing seamless integration with NVIDIA's GPUs, LPUs, networking, and storage platforms while leveraging NVIDIA's rich technology stack and global supply chain ecosystem.

The partnership will also see the two companies transform the world's communications networks into AI infrastructure through NVIDIA Aerial AI-RAN (for 5G/6G), and advance world-class networking for AI, including advanced optical interconnect solutions and silicon photonics technology.

NVIDIA Founder and CEO Jensen Huang said: “The inflection point for inference has arrived. Token generation demand is skyrocketing, and AI factories are being built around the world at a breakneck pace. Our collaboration with Marvell will enable customers to leverage NVIDIA’s AI infrastructure ecosystem to scale and build specialized AI computing.”

Marvell Chairman and CEO Matt Murphy said: “Our expanded partnership with NVIDIA is a testament to the growing importance of high-speed interconnects, optical interconnects and high-performance infrastructure in scaling AI. Marvell’s leadership in high-performance analog, optical DSP, silicon photonics and custom silicon, combined with NVIDIA’s expanding AI ecosystem, will be brought together through NVLink Fusion to enable customers to build scalable and efficient AI infrastructure.”

This announcement is a partial translation of the press release issued on Tuesday, March 31, 2026, U.S. time.