iPronics raises €20 million for optical networking technology targeting AI datacentres
The company says it will use the funding to accelerate the deployment of its Optical Networking Engine, which it says offers on-the-fly topology adaptation and 1000 times faster reconfiguration compared with other optics-based approaches
iPronics, a deep tech start-up in software-defined photonic engines, has announced it has raised a €20 million series A funding round. The company says this funding will accelerate the deployment of its Optical Networking Engine (ONE), in AI datacentres, enabling fast, scalable and high-bandwidth communication for energy-efficient AI. This investment is led by Triatomic Capital, with participation from Fine Structure Ventures and Bosch Ventures, Amadeus Capital Partners and Criteria Venture Tech.
“iPronics technology will enable larger GPU domains with fast optical interconnects, increasing training compute capacity and providing low latency inference,” said Christian Dupont, CEO of iPronics. “This investment will accelerate our Optical Circuit Switch market deployment, allowing customers to build next-gen AI infrastructure.”
Current AI datacentres use electronic switches that must be replaced as bandwidth needs to grow, limiting GPUs per server and creating fixed infrastructure that hinders efficient GPU utilisation. Optical networking can overcome these challenges.
According to iPronics, its ONE technology offers an optically switched fabric for AI architectures, enabling on-the-fly topology adaptation and extending programmability to physical layer connections. The company says it provides 1000 times faster reconfiguration compared with other optics-based approaches, lossless operation, lower cost-per-port, and higher reliability due to its solid-state chip design.
Peter Zhou, general partner at Triatomic Capital, said: “iPronics has the unique technology and team to advance photonics adoption in datacentres and AI clusters. We believe network fabrics operating entirely at the speed of light are essential to the next-generation compute stack. We are excited to support iPronics’ efforts to accelerate next-gen AI infrastructure performance.”
