The advent of 1-bit large language models (LLMs), led by BitNet b1.58, has spurred interest in ternary LLMs. Despite this, research and practical applications focusing on efficient edge inference for ternary LLMs remain scarce. To bridge this gap, we introducethis http URL, an inference system optimized for BitNet b1.58 and ternary LLMs. Given that mixed-precision matrix multiplication (mpGEMM) constitutes the bulk of inference time in ternary LLMs,this http URLincorporates a novel mpGEMM library to facilitate sub-2-bits-per-weight, efficient and lossless inference. The library features two core solutions: Ternary Lookup Table (TL), which addresses spatial inefficiencies of previous bit-wise methods, and Int2 with a Scale (I2_S), which ensures lossless edge inference, both enabling high-speed inference. Our experiments show thatthis http URLachieves up to a 6.25x increase in speed over full-precision baselines and up to 2.32x over low-bit baselines, setting new benchmarks in the field. Additionally, we expand TL to element-wise lookup table (ELUT) for low-bit LLMs in the appendix, presenting both theoretical and empirical evidence of its considerable potential.this http URLis publicly available atthis https URL, offering a sophisticated solution for the efficient and practical deployment of edge LLMs.
View on arXiv@article{wang2025_2502.11880,
title={ Bitnet.cpp: Efficient Edge Inference for Ternary LLMs },
author={ Jinheng Wang and Hansong Zhou and Ting Song and Shijie Cao and Yan Xia and Ting Cao and Jianyu Wei and Shuming Ma and Hongyu Wang and Furu Wei },
journal={arXiv preprint arXiv:2502.11880},
year={ 2025 }
}