Starting the Era of 1-bit LLMs – With Microsoft Research

by Brian Wang

BitNet is paving the way for a new era of 1-bit Large Language Models (LLMs). In this work, we introduce a 1-bit LLM variant, namely BitNet b1.58, in which every single parameter (or weight) of the LLM is ternary {-1, 0, 1}.</a> It matches the full-precision (i.e., FP16 or BF16) Transformer LLM with the same model size and training tokens in terms of bo…