1The University of Hong Kong 2 Beihang University 3ETH Zürich
Pretrained large language models (LLMs) exhibit exceptional general language processing capabilities but come with significant demands on memory and computational resources. As a powerful compression technology, binarization can extremely reduce model weights to a mere 1 bit, lowering the expensive computation and memory requirements. However, existing quantization techniques fall short of maintaining LLM performance under ultra-low bit-widths. In response to this challenge, we present BiLLM, a groundbreaking 1-bit post-training quantization scheme tailored for pretrained LLMs. Based on the weight distribution of LLMs, BiLLM first identifies and structurally selects salient weights, and minimizes the compression loss through an effective binary residual approximation strategy. Moreover, considering the bell-shaped distribution of the non-salient weights, we propose an optimal splitting search to group and binarize them accurately. BiLLM achieving for the first time high-accuracy inference (e.g. 8.41 perplexity on LLaMA2-70B) with only 1.08-bit weights across various LLMs families and evaluation metrics, outperforms SOTA quantization methods of LLM by significant margins. Moreover, BiLLM enables the binarization process of the LLM with 7 billion weights within 0.5 hours on a single GPU, demonstrating satisfactory time efficiency.
torch
: tested on v2.0.1+cu117transformers
: tested on v4.35.0 (the LLaMa integration currently requires a main install from source and sentencepiece
)datasets
: tested on v2.14.6huggingface-hub
: tested on v0.16.4All binarization processes and experiments were run on a single 80GB NVIDIA A100. However, all the process can also be conducted on a single 24GB NVIDIA 3090 Ti when the model's parameter is under 70B.
python3 run.py facebook/opt-6.7b c4 braq --blocksize 128 --salient_metric hessian
python3 run.py meta-llama/Llama-2-7b-hf c4 braq --blocksize 128 --salient_metric hessian
or
python3 run.py huggyllama/llama-7b c4 braq --blocksize 128 --salient_metric hessian
python3 run.py lmsys/vicuna-7b-v1.5 c4 braq --blocksize 128 --salient_metric hessian
We further evaluated BiLLM on 7 zero-shot dataset to give extensive insight on binarization LLMs
BiLLM achieve superior perplexity performance on Wikitext2 datasets within only an average of 1.10 bit-width weights Vicuna families (instruction fine-tune models).
GPTQ: Accurate Post-training Compression for Generative Pretrained Transformers
PB-LLM: Partially Binarized Large Language Models
AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration
If you find BiLLM is useful and helpful to your work, please kindly cite this paper:
@article{huang2024billm,
title={BiLLM: Pushing the Limit of Post-Training Quantization for LLMs},
author={Huang, Wei and Liu, Yangdong and Qin, Haotong and Li, Ying and Zhang, Shiming and Liu, Xianglong and Magno, Michele and Qi, Xiaojuan},
journal={arXiv preprint arXiv:2402.04291},
year={2024}
}