Lightmatter raises $155 million to develop photonic computing – Enerzine

Lightmatter, the Boston-based startup developing silicon photonics hardware for generative artificial intelligence (AI) and high-performance computing applications, announces it has raised an additional $155 million in funding. Capital risk.

Support for the “C-2 Series” led by Google Ventures and Viking Global Investors follows a similarly large initiative earlier this year, bringing total funding raised to date to $420 million.

The funding will allow the company, which is currently heavily recruiting, to further expand its workforce and plans to open a new office in Toronto next year. Lightmatter already has locations in Boston and Mountain View, California.

Co-founder and CEO Nick Harris said: “Lightmatter is poised to be a key player in developing the next generation of computing systems that will advance AI innovation even further.” »

“With photonic technologies, Lightmatter continues to advance computing power despite growing challenges in power consumption and slower progress in transistor scaling. »

Erik Nordlander, general partner at Google Ventures, added: “Lightmatter harnesses the power of silicon photonics to address the challenge of generative AI, eliminating performance bottlenecks, increasing bandwidth, and enabling the scale and scalability of AI models.” »

Lightmatter’s “Envise” photonic computing platform. The Boston-based company just raised an additional $155 million, bringing its total startup backing to $420 million and $1.2 billion. Valuation of the company. Image: Light matter.

Photonic computing

Among the products Lightmatter introduced earlier this year is the “Passage” photonic interconnect, which offers a significant improvement (40x) in interconnect density and enables parallel integration of silicon photonics and electronics.

Attaching fiber optics to chips is extremely expensive », emphasizes the company. “Passage has built-in photonic connections and no fibers need to be connected, which means less complexity. »

“By eliminating the need to run long cables between computer chips and optical transceivers, Passage achieves 5x more energy-efficient communications with 200x more bandwidth – reducing operating costs for high-performance computing systems. »

Additionally, the “Envise” hardware is described as “the world’s first photonic computing platform.” Lightmatter claims it is capable of running the largest neural networks developed to date while consuming a relatively low 3kW of power.

synthetic

Lightmatter is redefining the boundaries of artificial intelligence and high-performance computing with its innovative approach to photonic computing. With strong funding and an experienced management team, the company is well positioned to continue pushing the boundaries of technology and delivering revolutionary solutions in AI.

For better understanding

What is Photonic Computing?

Photonic computing is a technology that uses photons (light particles) to perform calculations, rather than electrons as in traditional computing. This helps increase the speed and efficiency of computer systems.

What is the C-2 series?

Series C-2 is a financing step for fast-growing companies. It follows Series C and is generally used to fund large-scale expansions.

What is Passage and Envise?

Passage is a photonic connection developed by Lightmatter that improves connection density. Envise is described as the world’s first photonic computing platform.

What is generative AI?

Generative AI is a form of artificial intelligence that uses algorithms to create new data from existing data.

What advantages does photonic computing offer?

Photonic computing offers several advantages, including faster computing speed, lower power consumption, and the ability to process larger amounts of data.

References

Source: Lightmatter

[ Rédaction ]