Lumai

2024 has been an eventful year for the Lumai team as we continue to push the boundaries of #AI acceleration using 3D optics. Here are some of the highlights: 🎉 Winning ‘Best Overall Technology’ at the Open Compute Project Foundation Future Technologies Symposium, as well as James Spall and Xianxin Guo being selected for the #Photonics100 and Royal Academy of Engineering’s Shott Accelerator. 🎉 Welcoming Howard Winter, Ivan Evans, Huiying Lan, Priya Gupta, and Benjamin McCann to our growing team. 🎉 Joining the OCP community and sharing insights at the AI Hardware Summit and SC Conference Series. A huge thank you to our team, partners, and supporters for making this year unforgettable! Read more in our latest blog: https://lnkd.in/e_qXxvmg

"Once the current hype subsides, AI’s broad technical application will mean a big focus on costs—specifically the cost of inference, which could make or break the business case for AI deployment."   As our CEO Tim Weil explains, reducing the cost of inference is crucial to unlocking AI’s full potential.   At Lumai, our 3D optics delivers #AI acceleration that not only pushes the boundaries of performance but also reduces both capital and operational costs.   Read more in IT Brief here: https://lnkd.in/eD3PKudA

Lumai has been featured in EE Times | Electronic Engineering Times to discuss our development of an AI accelerator using 3D optics to enable the lowest cost data centre inference. Our optical matrix-vector multiply accelerator offers a new approach to AI compute, processing inference workloads four times faster than today’s GPUs while significantly reducing power consumption and total cost of ownership by up to 90%. As Phillip Burr, Head of Product, explains: “Power consumption in the data centre is threatening the future of AI inference at scale. By leveraging free-space optics, we can reduce energy usage, cooling demands, and overall costs—key concerns for both businesses and the environment.”💡 With this approach, we’re tackling the limitations of silicon-based solutions, providing an efficient alternative that balances performance and sustainability. https://lnkd.in/eCCimw8a

Lumai has been featured in Intel Ignite's new report on the ‘State of UK Deep Tech 2024’, developed in collaboration with Beauhurst. As part of the first Intel Ignite London cohort, we’re proud to contribute to the UK’s thriving deep tech #startup ecosystem and drive innovation in #AI acceleration by using 3D optics. Read the full report below. 👇

Yesterday TechRound announced the winners of its #TechRound100, and Lumai has been recognised as one of the most innovative companies in the UK and Europe. As demand for increased #AI performance in #datacentres continues to grow, the industry is turning to transformative solutions. TechRound selected Lumai for our groundbreaking work in reducing the cost of AI by using a 3D optics-based AI processor. The AI accelerator meets this challenge by delivering a significant leap in AI performance while reducing both capital and operational costs. Want to learn about how Lumai is developing the world’s fastest and most energy-efficient AI datacentre accelerator? Read more: https://lnkd.in/es9h9qHd 

From taking home the best paper award at the Open Compute Project Foundation Global Summit in San Jose to celebrating one of our co-founders’ inclusion in the prestigious #Photonics100, it’s been a whirlwind of achievements for the Lumai team in recent months.   But we're just getting started! 🚀   Next stop: #SC24 in Atlanta!   If you're attending SC24 and are curious about how optics can accelerate AI with incredible efficiency and at a fraction of the total cost of ownership, don't miss the chance to connect with our Head of Product, Phillip Burr.

Lumai has been featured as one of Sifted’s ‘7 Oxford-based startups to watch according to investors.’   Selected by Alun Williams, Investment Director at Parkwalk Advisors, we’ve been recognised for our role in creating #AI acceleration with far greater compute speed and energy efficiency than traditional silicon processors.   Read the full line up, curated by investors including Octopus Ventures and Kindred Capital VC, below.   https://lnkd.in/ecNjxxP8

“Improving the energy efficiency of datacentres and processing hardware is incredibly important as the demand for AI continues to accelerate, and our 3D optical technology holds the answer to this challenge.” Our co-founder, James Spall, has been recognised in the Electro Optics Photonics100 list for 2025. This accolade highlights James' outstanding contribution to the field of optics, including the development of Lumai’s core optical technology, which is enabling AI acceleration at scale, during his PhD at the University of Oxford. Learn more below: https://lnkd.in/emrCp_S3 #Photonics100 #AI #Innovation 

Have you seen the news? The Open Compute Project Foundation (OCP) has crowned Lumai the winner of the ‘Best Overall Technology’ award at the Future Technologies Symposium 2024.   Lumai was recognised for addressing the limitations of AI compute through its unique 3D optics, to create technology which will achieve:   💡50-fold AI acceleration performance 💡Reduced costs by 90% 💡Energy consumption to only 10% of that currently needed by AI accelerators   Phillip Burr, Head of Product at Lumai, commented: “The industry recognises that the current trajectory of consuming ever-increasing energy to service AI is not sustainable. We all know that we need a different approach, and this award from OCP recognises the potential of optical compute to help solve this problem.”   Learn more:

On Friday 18th October, our co-founder, James Spall, will be speaking at the ZERO Founders Network - University of Oxford event on 'Unearthing the impact of #AI on the energy transition'.   James will share insights on his founding journey at Lumai, as well as details on how our ground-breaking 3D optics technology is solving the industry’s AI performance and sustainability challenges.   Find out more and register: https://lnkd.in/eB6d7ShR