ONNX

ONNX (Open Neural Network Exchange) is an open standard for representing machine learning models through a common set of operators and a shared file format, so models can move freely across frameworks, tools, runtimes, and compilers. It lets developers train in their preferred framework — such as PyTorch or TensorFlow — and deploy without being locked into a specific inference engine, while ONNX-compatible runtimes unlock hardware-level optimization. ONNX is a free, open-source community project governed as a graduate project of the Linux Foundation's LF AI & Data. It is widely used to standardize and accelerate model deployment across platforms.