As of now, the new AI-Xtension release 6.2.4 supports neural networks in the ONNX (Open Neural Network Exchange) format.

Another milestone for the NeuroCheck application software: In its latest version, the plug-in supports the general ONNX format for AI models, which are supported by a variety of AI software providers. With ONNX, AI developers can exchange models between various tools and select the combination of tools that works best for them. ONNX allows training of models in one framework and then transferring them to another framework for application. Thus, the use of TensorFlow or Keras is no longer absolutely necessary to create, train or maintain AI models for NeuroCheck. Users can stick to their preferred framework and just save their networks in ONNX format. Among others, the following frameworks are available: Caffe2, Matlab, MXNet, PyTorch, PaddlePaddle and Cognitive Toolkit. In addition, there are interfaces for many more commonly used frameworks and libraries.

NeuroCheck AI-Xtension ONNX Support New Release (IMage © NeuroCheck)

A sample project in demo mode* and a detailed white paper are available on the NeuroCheck website to get you started.

Our Support Team will be happy to answer your questions concerning the NeuroCheck AI-Xtension. You can reach them at

  • Phone +49 7146 8956-40
  • E-mail
  • NEW: ONNX Support

* Demo mode: Without a valid license, the check functions of
the plug-in will return an error message after every tenth execution.


Date: 15-10-2020

« Back to news overview


Copyright notice for the photos used on this page:
Images © NeuroCheck