Valencian Startup iPronics Raises $125M for AI Optical Switching 2026

iPronics, a Valencian startup founded in 2019 as a spin-off from the Universitat Politècnica de València, has raised $125 million in a Series B funding round to advance its optical switching technology for AI data centers. The round was co-led by Maverick Silicon and Light Street Capital, with participation from Nvidia, Bosch Ventures, Triatomic Capital, Catalight Capital, and the European Innovation Council fund. According to Reuters, the company has now accumulated $177 million in total funding.

This investment highlights a critical but often overlooked layer of the AI stack: the infrastructure that enables thousands of chips to communicate efficiently within data centers. As AI clusters grow, the ability to reconfigure networks quickly and reliably becomes essential to maintaining performance.

Why Optical Switching Matters for AI Data Centers

Modern AI data centers are built on massive clusters of GPUs that must exchange data at high speeds. Traditional electrical switches can become bottlenecks as cluster sizes increase, leading to latency and reduced throughput. Optical switching, which uses light to route signals, offers higher bandwidth and lower power consumption.

iPronics' technology is designed to make these optical networks reconfigurable, allowing data centers to adapt to changing workloads and recover from failures without significant downtime. This is particularly important for AI training and inference, where even brief interruptions can cost time and money.

What iPronics Brings to the Table

The company's photonic switching solutions are based on silicon photonics, which integrates optical components onto chips. This approach aims to reduce the cost and complexity of deploying optical switches in data centers. Daniel Pérez-López, founder and CTO of iPronics, explained to Reuters that the goal is to provide a "programmable optical fabric" that can be reconfigured on the fly.

With the new funding, iPronics plans to accelerate product development and scale its manufacturing. The involvement of Nvidia, a dominant player in AI hardware, signals strong strategic interest in the technology.

Key Investors and Their Roles

• Maverick Silicon and Light Street Capital co-led the round, bringing deep expertise in deep tech and growth-stage investments.
• Nvidia participated, likely to ensure its GPU clusters can leverage advanced optical networking.
• Bosch Ventures and Triatomic Capital add industrial and strategic backing.
• Catalight Capital and the European Innovation Council fund support European deep tech innovation.

How iPronics Fits into the AI Infrastructure Landscape

While much attention goes to GPU makers like Nvidia and model developers like OpenAI, the networking layer is equally critical. Companies such as Broadcom, Marvell, and Cisco dominate traditional data center networking, but optical switching startups are emerging to address the specific needs of AI workloads.

iPronics competes with other photonic switching players, but its focus on programmability and integration with silicon photonics could give it an edge. The backing from Nvidia suggests that the technology may be integrated into future AI systems.

What This Means for the AI Industry

The investment in iPronics underscores a broader trend: as AI models grow, the infrastructure supporting them must evolve. Optical switching could become a standard component of AI data centers, enabling faster and more efficient communication between GPUs.

For Nvidia, backing iPronics is a strategic move to ensure that its GPU clusters are not bottlenecked by networking. It also positions Nvidia to influence the development of optical standards. The background to this story is set out in Mistral Raises €3 Billion to Build AI Infrastructure and Challenge US Giants. A related thread runs through UK Startup Unit1 Raises $20m for Digital Avatar Concerts. Earlier coverage in Jeff Bezos and UK Government Invest in CuspAI Startup shows how the argument developed.

Challenges and What to Watch Next

Despite the promise, optical switching faces challenges in cost, reliability, and manufacturing scale. iPronics will need to prove that its technology can be deployed at scale in demanding data center environments. A closer look at the same subject appears in Jeff Bezos Invests in AI Startup CuspAI for Materials Discovery. Anyone following the wider picture can start with Jeff Bezos & UK Government Invest £2bn in AI Startup CuspAI. The same theme surfaces in Best VC Firms for AI Infrastructure and Agentic AI Startups in 2026, which takes a different angle.

Investors will be watching for pilot deployments and partnerships with major cloud providers. The company's ability to leverage Nvidia's ecosystem could accelerate adoption. Context on how this played out before sits in Best European Cities for Quality of Life and Cost: Malaga and Valencia. A companion piece, Business: latest articles, approaches the same subject from another angle.