Coherent Launched PhotonLink Platform for AI Infrastructure

The optics platform targets next-generation connectivity by integrating Indium Phosphide lasers with silicon photonics.

Updated on Oct. 5, 2026 in Quantum Computing

Isometric editorial illustration of a hexagonal grid of silicon prisms with stylized light beams, representing optical data transmission for AI.
Coherent launched its PhotonLink platform, integrating Indium Phosphide lasers with silicon photonics to boost bandwidth in AI infrastructure projects. AI Illustration. Upload story photo >

Live Poll

Is now a good time to invest in companies building AI hardware infrastructure?

Coherent has launched its PhotonLink platform, a suite of optics technology designed specifically for AI data infrastructure. The announcement includes over 25 active customer engagement projects.

Why it matters

As AI clusters demand higher bandwidth, integrating photonics directly into processing stacks has become a critical bottleneck for scaling infrastructure. This platform aims to address that demand by bridging silicon-based computing with high-speed optical data transmission.

The platform integrates Indium Phosphide lasers with silicon photonics and optical modules to facilitate data movement. Coherent has reached a manufacturing scale of 300 million Indium Phosphide lasers and is currently expanding its 6-inch fabrication capacity.

The players

Coherent

A provider of materials, networking, and laser technology that supplies components for data center optics and AI hardware infrastructure.

The details

PhotonLink uses Indium Phosphide — a semiconductor material with superior light-emitting properties compared to silicon — to generate optical signals. By combining these lasers with silicon photonics, which use light to move data across chips at high speeds, the platform reduces latency in AI clusters. Coherent is deploying this through Co-Packaged Optics (CPO) and Near-Packaged Optics (NPO), methods that place optical components closer to the processor to minimize distance and power consumption.

Timeline

  1. Q4 2026: Revenue ramp is expected for the platform.

The Tech Race

Coherent is competing in the high-stakes effort to replace copper-based electrical interconnects with light-based optical signaling in data centers. The development of CPO and NPO architectures represents a fundamental transition in how AI computing clusters scale their bandwidth requirements.

The platform is currently in the engagement phase, meaning it will initially impact large-scale data center operators and AI infrastructure providers. Widespread adoption and tangible impacts on computing speeds are expected to follow the projected revenue ramp in Q4 2026.

The takeaway

The transition to optical interconnects is the next necessary evolution for scaling power-intensive AI clusters. Watch for the revenue ramp in Q4 2026 as a leading indicator of how quickly these optics solutions are actually being adopted in commercial data center environments.

Further reading

For more background on the integration of hardware and light-based data transmission, see Quantum Computing.

Source note: This article includes information reported by Metal.

Live Poll

Is now a good time to invest in companies building AI hardware infrastructure?