mesa-teflon-delegate - 25.0.2-1 main

TensorFlow Lite delegate which can make use of NPUs to accelerate ML
inference. It is implemented in the form of a external delegate, a shared
library which the TensorFlow Lite runtime can load at startup.