Lumai introduces optical prefill server for disaggregated inference
The Iris Server uses optical matrix multipliers to reduce energy by 10x compared to silicon, targeting the prefill pool.
See the latest news and media coverage for Lumai. We track all announcements, press releases, and industry mentions in real time, all in one place.
3D optical computing AI accelerator developer
lumai.aiLast updated
In short: Lumai launched its Iris server, the world’s first optical computing system for real-time billion-parameter LLM inference, reducing energy.
The Iris Server uses optical matrix multipliers to reduce energy by 10x compared to silicon, targeting the prefill pool.
The post highlights the need for purpose-built optical computing to address massive energy and cost demands.
Optics moves from telecom to data centers to AI accelerators, culminating in Lumai Iris optical AI servers for LLM inference.
It runs real-time billion-parameter LLM inference with 90% lower energy, higher throughput, and lower cost than GPUs. Born from Oxford research for the inference era.
Lumai has unveiled what it describes as the world's first optical computing system capable of running a billion parameter large language model in real time...
Lumai, the optical compute company addressing scalable AI, today announced its Lumai Iris inference server – the world’s first optical computing system to successfully run...
Dr. Xianxin Guo, CEO and Co-Founder of Lumai, is a physicist and deep-tech entrepreneur specializing in optical computing and AI hardware, with a PhD in...
The future of artificial intelligence computing is at a turning point, and Lumai—a cutting-edge startup born out of the University ...
Track Lumai and your other target companies to get real-time alerts and weekly summaries delivered straight to your inbox.
Browse news for competitors to Lumai and other trending companies.