CyberTAN unveils 102.4 Tbps liquid cooled Ethernet switch blueprint

Oct 14, 2026
CyberTAN has introduced its Pallas Series switch platform blueprint, which combines Marvell Teralynx T100 silicon with direct chip liquid cooling for AI data centers.
CyberTAN unveils 102.4 Tbps liquid cooled Ethernet switch blueprint
Image: CyberTAN via PR Newswire

CyberTAN unveils its Pallas Series liquid cooled Ethernet switch platform blueprint for AI data centers, the company said in a press release. The architecture supports the 102.4 Tbps Marvell Teralynx T100 switch chip and 1.6T Ethernet connectivity.

The platform combines the 3 nm T100 silicon with direct chip liquid cooling. CyberTAN says the design can manage more than 1,000 watts of concentrated heat while supporting dense optical connections.

A prototype at the 2026 OCP Global Summit includes the chassis, optical housing, internal cabling, temperature sensors and a coolant distribution unit. The demonstration uses 40 W OSFP ZR and ZR+ optics with liquid entering at 45 degrees Celsius, and CyberTAN reports a power usage effectiveness measurement of 1.1. The platform was developed in collaboration with Marvell Technology, Inc..

We hope you enjoyed this article

Consider subscribing to one of our newsletters like Silicon Brief or Daily AI Brief.

Also, consider following us on social media:

AI Chips & Datacenters AI Brief AI Brief (X)
Free newsletter

Silicon Brief

Weekly coverage of AI hardware developments including chips, GPUs, cloud platforms, and data center technology.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.