Deep Learning Classification of Text

I have classified books into 5 language levels and collected about 100 books per language level. Now I need someone to build me a machine learning model that can predict language level for text of arbitrary length. Would prefer HuggingFace implementations, but am open to suggestions.