SPS 2025, Day 1: Beckhoff Live + Interactive, Nov 25, 2025

The TF3810 TwinCAT 3 Function is a high-performance execution module (inference engine) for trained neural networks.

The TF3800 TwinCAT 3 Function is a high-performance execution module (inference engine) for trained, conventional machine learning algorithms.

The TF3820 TwinCAT 3 Machine Learning Server is a high-performance service for executing trained AI models with the option of using hardware accelerators.

The TwinCAT 3 Machine Learning Server includes a connection to a local client as standard (local TwinCAT runtime). If (possibly further) TwinCAT runtimes need remote access to a TwinCAT 3 Machine Learning Server, these runtimes must each be equipped with a license for the TF3830 TwinCAT 3 Machine Learning Client.

TE3852 TwinCAT 3 MLC Signals and Time Series is an extension package for the basic TE3850 TwinCAT 3 Machine Learning Creator web application. This extension enables creation of AI models for signals and time series, e.g. classification, anomaly detection, and forecasting.

TE3851 TwinCAT 3 MLC Computer Vision is an extension package for the basic TE3850 TwinCAT 3 Machine Learning Creator web application. This extension allows AI models to be created for image processing, e.g. image classification, anomaly detection, and object recognition.

TE3860 TwinCAT 3 MLC Resource Pack is an extension package for the basic TE3850 TwinCAT 3 Machine Learning Creator web application. If more computing time is required, for example for training further AI models, additional computing time can be flexibly obtained via this extension package.

The TwinCAT 3 Machine Learning Creator automatically creates AI models based on data sets. These AI models can be optimized in terms of their accuracy and latency to ensure they run efficiently on Beckhoff Industrial PCs with TwinCAT products. The generated models can also be used as standardized ONNX models beyond the Beckhoff product range. For AI model execution with TwinCAT products, a PLCopen XML file with IEC 61131-3 code is created in addition to the model, which describes the complete AI pipeline and can be imported seamlessly into TwinCAT.

TwinCAT 3 CoAgent for Engineering is the intelligent AI assistant that provides efficient support for PLC development, I/O configuration, and human-machine interface creation. Requirements can be formulated in normal language and TwinCAT CoAgent for Engineering suggests suitable code and configurations, taking existing project structures into account. Flexible mode selection enables custom support based on the task at hand. In addition, TwinCAT CoAgent for Engineering offers direct access to the documentation in the Beckhoff Information System and provides context-related recommendations for accelerated high-quality project implementation.

TwinCAT 3 CoAgent for Operations is the intelligent AI assistant that provides efficient support in machine operation, maintenance, and system diagnostics. Operating and maintenance tasks are described in natural language. The CoAgent then provides specific recommendations for action, diagnostics, or direct system commands based on the current machine and process data.