Q.ANT released the world's first open-source SDK for photonic computing, enabling developers to write applications accelerated by photonic processors.
Q.ANT today debuted its open-source software development kit for photonic computing, making the programming tools for a photonic processor publicly available for the first time. Programmers now have everything they need to develop applications for Q.ANT's Native Processing Units (NPUs), enabling them to compute with light instead of electricity. The SDK includes APIs for C and Python, sample applications, documentation, and a simulation backend that runs on a standard CPU, allowing developers to build and validate applications on their own systems before running them on real photonic hardware.
The SDK's examples draw on Q.ANT's published reference applications, including an advanced neural network that achieves superior results with significantly fewer parameters—the source of Q.ANT's speed and energy-saving benefits.
In classical AI computing, most energy is consumed by constantly shuttling data between separate memory and processor. Q.ANT's photonic chip operates differently: computation happens directly in the optical domain, where non-linear functions are executed natively rather than approximated across many layers of transistors. More expressive networks can therefore be built with fewer parameters, requiring less data movement and compounding energy efficiency gains.
"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," said Michael Förtsch, Founder and CEO of Q.ANT. "An ecosystem isn't created by hardware alone. It emerges when the software layer is open and others can build on it. This is the 'Linux moment' of photonic computing: the point at which a new possibility becomes a global platform."
In the coming months, Q.ANT will open access to its hardware through the cloud via commercial partner IONOS or through on-premise installations of its Native Processing Server. Developers building with the SDK today will be ready to take full advantage of these new computing technologies once hardware access becomes available.
The SDK includes examples comparing classical neural networks with advanced networks based on photonic native operations, benchmarked on parameter count, accuracy, functional approximation, and training loss. Developers can download the SDK, integrate the Python or C API, and begin work immediately—no knowledge of photonics or special hardware required. Those ready to run simulations on real hardware can access it via cloud or on-premises.
"The ever-increasing cost, complexity, and power needs of traditional advanced computing systems are opening the door to new and innovative methods for solving today's most challenging computation problems," said Bob Sorensen, Senior Vice President of Research and Chief Analyst for Quantum Computing at Hyperion Research. "Q.ANT's innovative photonic processor, coupled with hardware availability through a cloud access model, offers the advanced computing community an ideal way to explore the performance potential of these pioneering systems."
Q.ANT, founded in 2018, commercializes photonic accelerators for AI and high-performance computing, offering a scalable alternative to transistor-based systems. Its Native Processing Units perform mathematical operations directly on Thin-Film Lithium Niobate-based photonic integrated circuits, enabling energy-efficient, high-performance co-processing for complex computational tasks. The company secured Europe's largest Series A for photonic computing in 2025 and introduced its second-generation processor that same year, now installed at leading scientific computing centers.
The SDK is available on GitHub at github.com/Q-ANT-GmbH/qant_native_computing_toolkit and github.com/Q-ANT-GmbH/qant_native_computing_toolkit_examples.