Q.ANT launches photonic computing SDK
Open-source SDK gives developers access to programming and simulation tools for Q.ANT's TFLN-based photonic processors.
Q.ANT has launched an open-source software development kit (SDK) for its photonic computing platform, giving developers tools to build and test applications for its photonic processors.
The SDK includes C and Python APIs, example applications, documentation and a simulation backend that runs on a standard CPU. Developers can therefore develop and validate applications without access to dedicated photonic hardware before transferring them to Q.ANT's processors.
Q.ANT's Native Processing Units perform mathematical operations directly in the optical domain using thin-film lithium niobate (TFLN)-based photonic integrated circuits (PICs). The company is targeting applications including AI inference and training, scientific computing and image processing.
The SDK includes examples comparing conventional neural networks with architectures using Q.ANT's photonic operations, covering parameters, accuracy, functional approximation and training loss.
Q.ANT plans to provide access to its photonic hardware through the cloud via IONOS, as well as through on-premise Native Processing Server installations.
The SDK is available through Q.ANT's open-source repositories, with the company planning to expand its supported photonic functions as the platform develops.
