Q.ANT has released the first open-source software development kit (SDK) for photonic computing, giving developers a new way to build AI applications that run on light-based processors instead of traditional electronic chips. The toolkit includes an application programming interface (API) for common programming languages and a simulation backend, allowing developers to start building and testing applications without needing immediate access to Q.ANT's physical photonic hardware.

What Makes Photonic Neural Processors Different From Traditional AI Chips?

Q.ANT's approach represents a fundamental shift in how artificial intelligence computations happen. Instead of using transistors to approximate complex mathematical functions, the company's Native Processing Units (NPUs) perform calculations directly in the optical domain, using the properties of light itself. This direct execution minimizes the need for complex approximations and reduces energy demands compared to conventional processors.

The key advantage lies in efficiency, not just raw speed. Traditional AI chips consume enormous amounts of power moving data between memory and processors. Q.ANT's photonic approach reduces this data movement significantly. The company demonstrated this through an example neural network that achieved superior results using significantly fewer parameters than conventional designs, directly addressing a critical limitation in traditional AI architectures where energy consumption is heavily tied to data transfer.

How Can Developers Start Building Photonic AI Applications Today?

• SDK Access: The open-source toolkit provides developers with an API for popular programming languages and a CPU-based simulation backend, allowing them to build and validate applications without requiring immediate access to Q.ANT's photonic hardware.
• Cloud Partnerships: Q.ANT is collaborating with high-performance computing centers including the Leibniz Supercomputing Centre (LRZ) and Jülich Supercomputing Centre (JSC), as well as cloud provider IONOS, to expand access to its technology.
• Benchmarking Tools: The SDK includes embedded benchmarks that allow developers to directly compare classical and photonic network architectures based on metrics like accuracy, functional approximation, and training loss, providing concrete evidence of the technology's potential.

Michael Förtsch, Founder and CEO of Q.ANT, framed this moment as a pivotal turning point for the emerging field.

"Markets and ecosystems are created by applications, and that is why Q.ANT is working every day to put photonic computing into developers' hands as fast as possible," he stated.

Michael Förtsch, Founder and CEO of Q.ANT

The company's strategy reflects a broader trend in the photonic computing landscape, where software tools are seen as critical for driving adoption. By releasing an open-source SDK, Q.ANT is lowering the barrier to entry and fostering broader adoption across the developer community.

When Will Photonic Hardware Actually Become Available?

Q.ANT has already secured significant funding to scale its hardware development. The company, a 2018 spinout from TRUMPF, secured 62 million euros in Series A funding in July 2025, followed by an 80 million dollar investment from Duquesne Family Office in November of the same year. This capital has enabled the development of its second-generation photonic processor, the NPU 2, a 19-inch rack-mountable unit with PCIe connectivity. Customer shipments are scheduled for the first half of 2026.

The company has also made strategic hires to accelerate its mission. Recent additions include Joerg Behrend, a veteran from IBM, who joined as Vice President of Hardware, signaling a commitment to scaling hardware development and delivering on the promise of photonic computing.

"The ever increasing cost, complexity, and power needs of traditional advanced computing systems are opening the door to new and innovative methods for solving some of today's most vexing computation problems," noted Bob Sorensen, Senior Vice President of Research and Chief Analyst for Quantum Computing at Hyperion Research.

Bob Sorensen, Senior Vice President of Research, Hyperion Research

Sorensen added that Q.ANT's system, coupled with the new SDK and cloud access, "offers the advanced computing community an ideal way to explore the performance potential of Q.ANT's systems". This accessibility is crucial because photonic computing remains largely experimental, and developers need hands-on experience to understand how to leverage its unique advantages.

Sorensen

The release of this open-source SDK marks the first time developers have openly programmable access to a photonic processor. By democratizing access to this emerging technology, Q.ANT is positioning itself at the forefront of a potential paradigm shift in how AI computations are performed, moving from electricity-based transistors to light-based optical systems that could fundamentally reshape the efficiency and scalability of artificial intelligence workloads.