Latte is a high-performance DSL for deep neural networks. Our goal was to design a language that closely matched domain concepts without sacrificing performance.


Please note that Latte is prelease (alpha) software and this site is a work in progress. We are working hard on filling out this site with documentation, tutorials, and examples. Please report bugs using the Github issue tracker. For usage questions, visit the Gitter room.

If you use Latte for research, please use the following citation: