A more memory-efficient rewrite of the HF transformers implementation of Llama for use with quantized weights.

project main repository

Android application

Language Detection with Infinity-gram

Machine Learning / Natural Language Processing / Information Retrieval

learning cpp