Lumai is rethinking the hardware that runs AI by moving computation off silicon and into light. Its optical computing technology performs the matrix-multiplication operations at the heart of neural networks using beams of light traveling through three-dimensional space, sidestepping the power and bandwidth limits that constrain conventional GPU and silicon accelerators.
The company's pitch centers on the economics of AI inference, which is becoming the dominant cost in deploying large models. Lumai claims its approach can deliver large performance gains while using a fraction of the power of current solutions, with its flagship Lumai Iris system described as capable of running billion-parameter LLMs in real time, improving data-center sustainability in the process.
Founded in 2022 as a spinout from the University of Oxford, Lumai is based in the Oxford area of the UK. It was founded by a team of optical and computing researchers including Tim Weil, Xianxin Guo, Alex Lvovsky, Thomas Barrett, and James Spall.
In April 2025, Lumai raised over $10 million in a round led by Constructor Capital, with support from IP Group and new backers including PhotonVentures, Journey Ventures, LIFTT, and others. The funding is being used to grow the team, accelerate product development, and expand its presence in the US.
Lumai positions optical computing as a path to scalable, energy-efficient AI inference as model sizes and inference demand continue to grow.