VPTQ: Extreme Low-bit Vector Post-Training Quantization for Large Language Models Article Swipe

MAKE IMPACT

Yifei Liu , Jicheng Wen , Yang Wang , Shengyu Ye , Li Lyna Zhang , Ting Cao , Cheng Li , Mao Yang · YOU? · · 2024 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2409.17066

Scaling model size significantly challenges the deployment and inference of Large Language Models (LLMs). Due to the redundancy in LLM weights, recent research has focused on pushing weight-only quantization to extremely low-bit (even down to 2 bits). It reduces memory requirements, optimizes storage costs, and decreases memory bandwidth needs during inference. However, due to numerical representation limitations, traditional scalar-based weight quantization struggles to achieve such extreme low-bit. Recent research on Vector Quantization (VQ) for LLMs has demonstrated the potential for extremely low-bit model quantization by compressing vectors into indices using lookup tables. In this paper, we introduce Vector Post-Training Quantization (VPTQ) for extremely low-bit quantization of LLMs. We use Second-Order Optimization to formulate the LLM VQ problem and guide our quantization algorithm design by solving the optimization. We further refine the weights using Channel-Independent Second-Order Optimization for a granular VQ. In addition, by decomposing the optimization problem, we propose a brief and effective codebook initialization algorithm. We also extend VPTQ to support residual and outlier quantization, which enhances model accuracy and further compresses the model. Our experimental results show that VPTQ reduces model quantization perplexity by $0.01$-$0.34$ on LLaMA-2, $0.38$-$0.68$ on Mistral-7B, $4.41$-$7.34$ on LLaMA-3 over SOTA at 2-bit, with an average accuracy improvement of $0.79$-$1.5\%$ on LLaMA-2, $1\%$ on Mistral-7B, $11$-$22\%$ on LLaMA-3 on QA tasks on average. We only utilize $10.4$-$18.6\%$ of the quantization algorithm execution time, resulting in a $1.6$-$1.8\times$ increase in inference throughput compared to SOTA.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2409.17066

PDF

https://arxiv.org/pdf/2409.17066

OA Status

green

Related Works

10

OpenAlex ID

https://openalex.org/W4403785034

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4403785034

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2409.17066

Digital Object Identifier

Title

VPTQ: Extreme Low-bit Vector Post-Training Quantization for Large Language Models

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2024

Year of publication

Publication date

2024-09-25

Full publication date if available

Authors

Yifei Liu, Jicheng Wen, Yang Wang, Shengyu Ye, Li Lyna Zhang, Ting Cao, Cheng Li, Mao Yang

List of authors in order

Landing page

https://arxiv.org/abs/2409.17066

Publisher landing page

PDF URL

https://arxiv.org/pdf/2409.17066

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2409.17066

Direct OA link when available

Concepts

Vector quantization, Bit (key), Computer science, Quantization (signal processing), Training (meteorology), Artificial intelligence, Theoretical computer science, Algorithm, Physics, Computer security, Meteorology

Top concepts (fields/topics) attached by OpenAlex

Cited by

0

Total citation count in OpenAlex

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload