paper-with-me

Papers

Evaluating Quantized Large Language Models

2024-02-28 · Shiyao Li, Xuefei Ning, Luning Wang, Tengxuan Liu, Xiangsheng Shi, Shengen Yan, Guohao Dai, Huazhong Yang, Yu Wang

Post-training quantization (PTQ) has emerged as a promising technique to reduce the cost of large language models (LLMs). Specifically, PTQ can effectively mitigate memory consumption and reduce computational overhead in LLMs. To meet the requirements of both high efficiency and performance across diverse scenarios, a comprehensive evaluation of quantized LLMs is essential to guide the selection of quantization methods. This paper presents a thorough evaluation of these factors by evaluating the effect of PTQ on Weight, Activation, and KV Cache on 11 model families, including OPT, LLaMA2, Falcon, Bloomz, Mistral, ChatGLM, Vicuna, LongChat, StableLM, Gemma, and Mamba, with parameters ranging from 125M to 180B. The evaluation encompasses five types of tasks: basic NLP, emergent ability, trustworthiness, dialogue, and long-context tasks. Moreover, we also evaluate the state-of-the-art (SOTA) quantization methods to demonstrate their applicability. Based on the extensive experiments, we systematically summarize the effect of quantization, provide recommendations to apply quantization techniques, and point out future directions. The code can be found in https://github.com/thu-nics/qllm-eval.

📄 PDF Abstract BibTeX arXiv:2402.18158

Code (1)

thu-nics/qllm-eval 공식 구현 pytorch

Tasks

MambaQuantization

Methods 이 논문이 사용한 방법론

OPT OPT is a suite of decoder-only pre-trained transformers ranging from 125M to 175B parameters. The model uses an AdamW optimizer and weight decay of 0.1. It follows a linear…

Similar Papers 제목 키워드 기반

Quantization Hurts Reasoning? An Empirical Study on Quantized Reasoning Models

2025-04-07 · Ruikang Liu, Yuxuan Sun, Manyi Zhang, Haoli Bai 외

Recent advancements in reasoning language models have demonstrated remarkable performance in complex tasks, but their extended chain-of-thought reasoning process increases inference overhead. While quantization has been …

MathQuantization

Evaluating Quantized Large Language Models for Code Generation on Low-Resource Language Benchmarks

2024-10-18 · Enkhbold Nyamsuren

Democratization of AI is an important topic within the broader topic of the digital divide. This issue is relevant to LLMs, which are becoming popular as AI co-pilots but suffer from a lack of accessibility due to high c…

Code GenerationGPUQuantization

Low-Bit Quantization Favors Undertrained LLMs: Scaling Laws for Quantized LLMs with 100T Training Tokens

2024-11-26 · Xu Ouyang, Tao Ge, Thomas Hartvigsen, Zhisong Zhang 외

We reveal that low-bit quantization favors undertrained large language models (LLMs) by observing that models with larger sizes or fewer training tokens experience less quantization-induced degradation (QiD) when applyin…

Quantization

Quantization Inflates Reasoning: Token Inflation as a Hidden Cost of Low-Bit Reasoning Models

2026-06-24 · Xinyu Lian, Walid Krichene, Beichen Huang, Masahiro Tanaka 외 arxiv

Quantization is widely used to reduce the inference cost of large language models, but its effect on reasoning models is not fully captured by final-answer accuracy or per-token latency. We show that low-bit post-trainin…

Mathematical ReasoningQuestion AnsweringCode Generation

INT2.1: Towards Fine-Tunable Quantized Large Language Models with Error Correction through Low-Rank Adaptation

2023-06-13 · Yuji Chai, John Gkountouras, Glenn G. Ko, David Brooks 외

We introduce a method that dramatically reduces fine-tuning VRAM requirements and rectifies quantization errors in quantized Large Language Models. First, we develop an extremely memory-efficient fine-tuning (EMEF) metho…

Language ModelingLanguage ModellingLarge Language ModelQuantization