
HyperLight offers TFLN Chiplet Platform
TFLN Chiplet Platform is HyperLight's thin-film lithium niobate photonic integrated circuit platform for high-bandwidth optical interconnect.
- Currently Offered

TFLN Chiplet Platform is HyperLight's thin-film lithium niobate photonic integrated circuit platform for high-bandwidth optical interconnect.
HyperLight's TFLN Chiplet Platform is a thin-film lithium niobate photonic integrated circuit platform designed for high-volume manufacturing. It combines TFLN's low-loss, high-bandwidth electro-optic properties with CMOS-like fabrication to deliver optical interconnect solutions for AI data centers, telecom networks, and emerging photonics applications.
The platform supports IMDD, coherent, and co-packaged optics use cases, offering known-good dies, custom development, and multi-project wafer runs. It is manufactured through partnerships with foundries such as UMC and Wavetek and targets energy-efficient scaling of optical I/O.


Passage is Lightmatter's photonic interconnect platform that uses 3D vertical integration to scale bandwidth density for AI training systems.
Passage is Lightmatter's photonic interconnect platform that uses 3D vertical integration to scale bandwidth density for AI training systems.



Photonic Fabric is Marvell's optical interconnect platform for chiplet-based, multi-rack AI scale-up networks.


TeraPHY is Ayar Labs' optical I/O chiplet that delivers high-bandwidth, low-latency, power-efficient interconnect for AI accelerators.


Optical I/O pairs Ayar Labs' TeraPHY engines and SuperNova light sources for co-packaged AI scale-up connectivity.


MicroLED optical interconnects that link chips over multi-core fiber for AI scale-up.


NVIDIA NVL is a rack-scale AI compute system architecture built around NVLink scale-up interconnect for large GPU clusters.
Photonic Fabric is Marvell's optical interconnect platform for chiplet-based, multi-rack AI scale-up networks.

TeraPHY is Ayar Labs' optical I/O chiplet that delivers high-bandwidth, low-latency, power-efficient interconnect for AI accelerators.

Optical I/O pairs Ayar Labs' TeraPHY engines and SuperNova light sources for co-packaged AI scale-up connectivity.

MicroLED optical interconnects that link chips over multi-core fiber for AI scale-up.

NVIDIA NVL is a rack-scale AI compute system architecture built around NVLink scale-up interconnect for large GPU clusters.
