Q.ANT, the Stuttgart-based pioneer in photonic processing, has announced the second close of its Series A round, bringing total funding to $80 million. Previously, the company closed nearly $72 million in the first tranche of its Series A round. This round marks the largest photonic computing raise in Europe.
The latest investment comes from Duquesne Family Office, the firm of billionaire investor Stanley F. Druckenmiller. The funds will drive commercialisation, expand operations in the U.S., and accelerate development of Q.ANT’s light-based processors. Q.ANT did not disclose the valuation in a conversation with TFN.
Duquesne Family Office joins current lead investors, including Cherry Ventures, UVC Partners and imec.xpand and other deep tech investors, including L-Bank, Verve Ventures, Grazia Equity, EXF Alpha of Venionaire Capital, LEA Partners, Onsight Ventures, and TRUMPF.
The company went on to state that, “This round shows Europe can scale deep-tech and sets a blueprint for next-generation, sustainable compute. Our mission remains clear: rethink computing from the ground up, harnessing light to deliver scalable performance for AI and HPC.”
What was the primary motivation for starting Q.ANT?
Q.ANT was founded by Dr Michael Förtsch, a physicist and former Max Planck researcher who led the spin-out from TRUMPF. The company is built on expertise in photonics, semiconductor manufacturing, and system architecture.
It was revealed that “We have been developing photonic chip technology since 2018. Since 2022, we have specifically been building AI accelerators because, even before AI became widely used, we foresaw an impending energy and performance problem as AI entered the mainstream market. Traditional CMOS technology is reaching its physical and economic limits, especially for AI and HPC. Our team saw an urgent market need for more efficient compute architectures, and believed light—not electricity—would be the answer.”
“We realised that we could perform the basic computations necessary for such AI workloads very efficiently on our chip. So we have focused on this technology because we see the greatest market opportunity here. This vision was informed by years of academic research and hands-on industry experience, as well as our extensive work with Thin Film Lithium Niobate, a material optimal for computing with light.”
What challenge does Q.ANT tackle?
Q.ANT tackles the fundamental limits of today’s compute infrastructure—energy use, performance, and cost. Digital computing is running out of steam: AI data centres are projected to consume up to 17% of US electricity by 2030, and operational costs are spiralling as a single GPU now draws as much power as a kitchen oven. Data centre energy use is forecast to rise by 160% this decade.
Traditional chips aren’t built for the complexity of modern AI. An 8-bit multiplication on a CMOS processor needs 1,200 transistors; on Q.ANT’s photonic chip, it takes just one optical element, delivering up to 30x higher energy efficiency. Even common operations like a Fourier transform, which require over a million transistors on silicon, are executed natively in light with a single component.
The industry is reaching physical and economic limits. Existing GPU racks demand extreme cooling and huge footprints, while Q.ANT’s passive-cooled servers enable higher density—cutting rack-level power consumption by up to 90x. And instead of billion-dollar new fabs, we industrialised our technology on a repurposed 1990s CMOS line in Germany at a fraction of the cost.
Competition: What sets Q.ANT apart?
The company stated, “Globally, several startups and research groups are pursuing photonic or hybrid processors, including Lightmatter (US), Celestial AI (US), Arago (France), and Oriole (Israel). Most focus on integrating optics for interconnects or are still at the demonstration phase.”
Competitors in the international context used a different material. After some time, they had to look for other applications because their processors did not meet the quality standards the market would accept. They did not reach the precision. Before Q.ANT, the maximum precision was 5 bits. Since we have been around, 16-bit is no problem. We also have the prospect of 32-bit.
They continued, “Q.ANT’s differentiation: we offer a commercially available, analogue photonic processor built on TFLN and a scalable production model. What we deliver today: a turnkey system that integrates seamlessly into a data centre. This is unique in the photonic computing industry. Nobody else has that.”
Q.ANT delivers photonic processors based on thin-film lithium niobate, allowing direct analogue computation with light. This approach skips the power-hungry translation to digital bits, solving non-linear equations natively—critical for AI. Unlike most “photonic” or quantum competitors still in R&D, Q.ANT’s first-generation products are easy to integrate into today’s data centres. Another unique aspect of our solution is that our chips are manufactured on a repurposed 90nm CMOS line in Germany, demonstrating scalability, further cost efficiency, and European chip sovereignty.
The benefits customers can get
As stated by the company, the advantages for their customers, including data centre operators and their customers, the AI developers, are as follows:
- The data centre operator has lower electricity costs – that’s an economic added value. Our servers come without water cooling and without installation costs; you don’t need to consider the special load capacity of your data centre floor: it’s a bit like installing servers in the past, just four screws and two cables.
- Software users realise that they can run certain applications faster than they used to be, too complex to program or took too long to calculate: e.g. physics simulations that used to take too long; or AI in the image area (image recognition, Image segmentation, text to image, etc). Our processor can be used to create training sets that were neither economically viable nor time-relevant in the past. Calling up libraries, executing code structures.
What’s next?
Over the next five to seven years, Q.ANT’s focus is to establish photonic computing as a new foundation for AI and high-performance computing. The company stated, “We’ll continue to push the limits of our Native Processing Server, increasing precision, speed, and density, with regular product advancements that keep us ahead of industry needs. We are confident that we can perform 50-60% of AI tasks faster and more efficiently. Our priority is to bring this into the hands of developers and into applications – the entire software ecosystem has to grow with it. This means, we need people to use it, to write code and develop applications on it. By making it easy for them – we are seamlessly integrated into the existing AI software stack and support programming languages like C++ and Python, we are confident to win them over.“
They further stated, “We’re expanding capacity at our pilot line in Stuttgart and have a model that can be replicated at other foundries worldwide—making advanced chip manufacturing more local, cost-effective, and resilient. Ultimately, we see ourselves as a processor company, not a manufacturer, so we plan to license our technology broadly as the market matures. For now, our technology is in the hands of early adopters and innovators who are already preparing for the next era of compute. By the end of this decade, our aim is for photonic co-processors to be standard in major data centres, working alongside CPUs and GPUs to deliver the performance and efficiency today’s workloads demand.”
“AI is pushing the limits of global resources – energy, hardware, and capital,” said Dr. Michael Förtsch, founder and CEO of Q.ANT. “At Q.ANT, we achieve performance through efficiency, not brute power alone, redefining how AI can scale. The Duquesne Family Office shares our conviction that sustainable computing will define the next era of progress.”
