Marvell Unveils 1.6T Silicon Photonics Light Engine for AI Networks

Marvell Technology has introduced its 1.6T silicon photonics light engine, designed to enhance AI network connectivity with lower power and reduced latency, as announced in a press release.

Marvell Technology, Inc. has introduced its 1.6T silicon photonics light engine, aimed at improving AI network connectivity by offering lower power consumption and reduced latency. Announced in a press release, the light engine was demonstrated at OFC 2025 in a 1.6T OSFP LPO module. This new product is part of Marvell's light engine family, which also includes a 6.4T light engine for co-packaged optics.

The 1.6T light engine supports eight channels of 200 Gbps PAM4 optical interconnect with standards-based signaling, providing a solution for mid-density rack-scale connectivity. It integrates a linear driver, TIA, and silicon photonics chip into a single package, simplifying system integration and accelerating time to market for AI server deployments.

Designed to address the limitations of passive copper interconnects, the light engine offers a low-power alternative with longer reach and lower latency. It is currently sampling to select customers, providing a rapid path to commercialization for ecosystem partners.

We hope you enjoyed this article.

Consider subscribing to one of several newsletters we publish like Silicon Brief.

Also, consider following our LinkedIn page AI Chips & Datacenters.

Subscribe to Daily AI Brief

Daily report covering major AI developments and industry news, with both top stories and complete market updates