Eliyan Secures $60 Million in Funding

Eliyan, the chiplet interconnect specialist, has successfully closed a $60 million funding round, with notable investors including Samsung Catalyst Fund, Tiger Global Management, Hynix, Intel Capital, Cleveland Avenue, and Mesh Ventures.

This recent funding round comes on the heels of Eliyan's $40 million Series A round in 2022. The infusion of capital will empower Eliyan to further concentrate on overcoming the challenges associated with designing and manufacturing advanced AI chips utilizing multi-die architectures within advanced packaging or standard organic substrates.

One of the key focuses of Eliyan is on die-to-die interconnect in chiplet-based designs, as well as addressing the escalating demands for memory capacity and bandwidth in AI chips through its innovative Universal Memory Interface (UMI).

The bi-directional interconnect approach of UMI is specifically designed to combat the "memory wall" obstacle encountered in large, multi-die designs. By facilitating a highly bandwidth-efficient connection to memory in both standard organic substrates and advanced packaging, UMI significantly enhances aggregate memory bandwidth per AI chip while reducing the die area required for memory interfaces.

Eliyan recently achieved a significant milestone with its NuLink PHY, which successfully taped out on TSMC's cutting-edge 3nm process. The NuLink PHY boasts impressive performance capabilities of up to 64Gbs per link, setting a new benchmark for performance-to-power ratio in the industry.

"This substantial investment underscores the confidence in our strategy of integrating multi-chip architectures to tackle the critical challenges of high costs, low yield, power consumption, manufacturing complexity, and size constraints," stated Eliyan's co-founder and CEO, Ramin Farjadrad. He further emphasized, "Our NuLink technology has reached commercial readiness with successful tape outs in the most advanced processes, and is finely tuned to deliver the essential high bandwidth, low latency, and low power features demanded by the market."