Lightmatter Inc., a startup building computing products that use light for computing, said today that it has raised $155 million in additional funding increasing the company’s valuation to $1.2 billion.

The funding was led by GV, the venture capital arm of Alphabet Inc., and Viking Global Investors, which is an extension to the company’s Series C financing round in May that raised $154 million. The new round brings the total raised by the company to more than $420 million and Lightmatter said it has doubled its headcount, which puts its team at around 150.

The company produces photonic technologies that change how computer chips perform calculations and communication by using light instead of electrical signals to transmit data. As chips become smaller and performance increases, challenges have risen in increasing performance and speed, since signals can bleed between the chips. This problem doesn’t exist when photons, or light, are used to propagate signals.

Using light to build computer chips and help them communicate isn’t a new process for the tech industry. Indeed, it has been suggested as a strategy get past what is called Moore’s law. Moore’s law is an observation that the number of transistors in computer chips doubles about every two years, but eventually physics will catch up — some contend it already has — and high temperatures or cost or other challenges would make it impossible to make smaller or more dense chips.

Lightmatter provides a full stack of photonics-based hardware and software solutions that allows for increasing performance, efficiency and power consumption for compute-intensive workloads, which the company says is particularly essential for high-performance activities such as artificial intelligence. With the popularity of large language model AI such as OpenAI’s ChatGPT and diffusion image models such as DALL-E and Midjourney, higher performance at lower cost is becoming more critical.

“Lightmatter is positioned to be a key driver in powering the next generation of computing systems that will further enable AI innovation,” said co-founder and Chief Executive Nick Harris. “Through photonic technologies, Lightmatter is ensuring the steady progress in computing performance continues, despite growing power consumption challenges and slowing progress with transistor scaling.”

The company provides products such as Passage, which is an optical interconnect device that allows computer chips to share data using photonics. This allows computer chips to share data with high bandwidth and extremely low latency that can be scaled up at low cost. According to the company, it can be scaled up to supercomputing scales and provides more than 100x more bandwidth than existing chip-to-chip interconnect solutions.

With Lightmatter’s Envise, a server architecture that combines electronics and photonics, customers have at their disposal a platform purpose-built for AI workloads. Envise is offered as a server blade form factor that offers up to 6.4 terabytes per second optical data interconnect for scaling up, 1 terabyte of DDR4 DRAM and 3 terabytes of solid-state memory.

The company also provides a software solution called Idiom, that permits developers to interface with standard deep learning frameworks and AI model formats to run machine learning on Lightmatter hardware. Using Idiom, developers do not need to change their development deep learning models written in PyTorch, TensorFlow or ONNX formats to have them run on Lightmatter computing platforms.

The company said that with the new funding, it will push forward with growth to meet the increasing demand for high-performance computing from AI companies, although it didn’t name any specific customers. Lightmatter also said that with its increased team size, it plans to open an office in Toronto in 2024 in addition to its Boston location.

Image: Lightmatter