Which of the following is NOT true about Databricks ML clusters?

Databricks ML clusters are designed to be flexible and powerful, supporting a variety of workloads. The assertion that they are limited to CPU processing only is not accurate. In fact, these clusters can utilize pre-configured GPU support, which allows users to leverage high-performance graphics processing units for computationally intensive tasks, such as deep learning. This capability is particularly beneficial for training deep learning models, which often require substantial processing power.

Additionally, these clusters come preloaded with popular machine learning libraries, such as TensorFlow and PyTorch, enabling data scientists and developers to quickly start their projects without needing to worry about the underlying infrastructure or library management. The availability of these libraries and the support for both CPU and GPU processing significantly enhance the usability and performance of Databricks ML for various machine learning tasks.