# ParetoQ: Improving Scaling Laws in Extremely Low-bit LLM Quantization

arXiv: https://arxiv.org/abs/2502.02631 · HTML: https://mlsanity.com/paper/2502.02631/

## Key facts

- Cited by 46 papers in the map (data through 14 Oct 2026).
- Builds on 46 of its 55 references that are in the map.
- First posted on arXiv 4 Feb 2025.

## Builds on (most cited first, 30 of 46)

- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [HellaSwag: Can a Machine Really Finish Your Sentence?](https://mlsanity.com/paper/1905.07830/) (2019)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [PIQA: Reasoning about Physical Commonsense in Natural Language](https://mlsanity.com/paper/1911.11641/) (2019)
- [BoolQ: Exploring the Surprising Difficulty of Natural Yes/No Questions](https://mlsanity.com/paper/1905.10044/) (2019)
- [Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answering](https://mlsanity.com/paper/1809.02789/) (2018)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale](https://mlsanity.com/paper/2208.07339/) (2022)
- [RACE: Large-scale ReAding Comprehension Dataset From Examinations](https://mlsanity.com/paper/1704.04683/) (2017)
- [XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks](https://mlsanity.com/paper/1603.05279/) (2016)
- [Crowdsourcing Multiple Choice Science Questions](https://mlsanity.com/paper/1707.06209/) (2017)
- [QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs](https://mlsanity.com/paper/2404.00456/) (2024)
- [Learned Step Size Quantization](https://mlsanity.com/paper/1902.08153/) (2019)
- [Quantizing deep convolutional networks for efficient inference: A whitepaper](https://mlsanity.com/paper/1806.08342/) (2018)
- [A White Paper on Neural Network Quantization](https://mlsanity.com/paper/2106.08295/) (2021)
- [SpinQuant: LLM quantization with learned rotations](https://mlsanity.com/paper/2405.16406/) (2024)
- [OmniQuant: Omnidirectionally Calibrated Quantization for Large Language Models](https://mlsanity.com/paper/2308.13137/) (2023)
- [QuIP: 2-Bit Quantization of Large Language Models With Guarantees](https://mlsanity.com/paper/2307.13304/) (2023)
- [QuIP#: Even Better LLM Quantization with Hadamard Incoherence and Lattice Codebooks](https://mlsanity.com/paper/2402.04396/) (2024)
- [LLM-QAT: Data-Free Quantization Aware Training for Large Language Models](https://mlsanity.com/paper/2305.17888/) (2023)
- [The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits](https://mlsanity.com/paper/2402.17764/) (2024)
- [SqueezeLLM: Dense-and-Sparse Quantization](https://mlsanity.com/paper/2306.07629/) (2023)
- [Atom: Low-bit Quantization for Efficient and Accurate LLM Serving](https://mlsanity.com/paper/2310.19102/) (2023)
- [The case for 4-bit precision: k-bit Inference Scaling Laws](https://mlsanity.com/paper/2212.09720/) (2022)
- [Bi-Real Net: Enhancing the Performance of 1-bit CNNs With Improved Representational Capability and Advanced Training Algorithm](https://mlsanity.com/paper/1808.00278/) (2018)

## Built on by (most cited first, 30 of 46)

- [Apple Intelligence Foundation Language Models: Tech Report 2025](https://mlsanity.com/paper/2507.13575/) (2025)
- [TorchAO: PyTorch-Native Training-to-Serving Model Optimization](https://mlsanity.com/paper/2507.16099/) (2025)
- [Scaling Law for Quantization-Aware Training](https://mlsanity.com/paper/2505.14302/) (2025)
- [Pushing the Limits of Block Rotations in Post-Training Quantization](https://mlsanity.com/paper/2601.22347/) (2026)
- [What Makes Low-Bit Quantization-Aware Training Work for Reasoning LLMs? A Systematic Study](https://mlsanity.com/paper/2601.14888/) (2026)
- [Hy-MT2: A Family of Fast, Efficient and Powerful Multilingual Translation Models in the Wild](https://mlsanity.com/paper/2605.22064/) (2026)
- [ML-SpecQD: Multi-Level Speculative Decoding with Quantized Drafts](https://mlsanity.com/paper/2503.13565/) (2025)
- [Tequila: Trapping-free Ternary Quantization for Large Language Models](https://mlsanity.com/paper/2509.23809/) (2025)
- [Unifying Block-wise PTQ and Distillation-based QAT for Progressive Quantization toward 2-bit Instruction-Tuned LLMs](https://mlsanity.com/paper/2506.09104/) (2025)
- [WaterSIC: Information-Theoretically (Near) Optimal Linear Layer Quantization](https://mlsanity.com/paper/2603.04956/) (2026)
- [GEMQ: Global Expert-Level Mixed-Precision Quantization for MoE LLMs](https://mlsanity.com/paper/2605.23078/) (2026)
- [Gated Relational Alignment via Confidence-based Distillation for Efficient VLMs](https://mlsanity.com/paper/2601.22709/) (2026)
- [Statistically-Lossless Quantization of Large Language Models](https://mlsanity.com/paper/2605.02404/) (2026)
- [Improving Quantization with Post-Training Model Expansion](https://mlsanity.com/paper/2503.17513/) (2025)
- [Unified Scaling Laws for Compressed Representations](https://mlsanity.com/paper/2506.01863/) (2025)
- [iFairy: the First 2-bit Complex LLM with All Parameters in ±1, ± i](https://mlsanity.com/paper/2508.05571/) (2025)
- [A Survey on the Green Development of Large Models: From Resource-Efficient Architectures to Hardware-Software Co-Design](https://mlsanity.com/paper/2607.09084/) (2026)
- [Characterization and Mitigation of Training Instabilities in Microscaling Formats](https://mlsanity.com/paper/2506.20752/) (2025)
- [Not All Bits Are Equal: Scale-Dependent Memory Optimization Strategies for Reasoning Models](https://mlsanity.com/paper/2510.10964/) (2025)
- [LC-QAT: Data-Efficient 2-Bit QAT for LLMs via Linear-Constrained Vector Quantization](https://mlsanity.com/paper/2606.10531/) (2026)
- [Optimal Post-Training Quantization Scales and Where to Find Them](https://mlsanity.com/paper/2606.10890/) (2026)
- [Vec-LUT: Vector Table Lookup for Parallel Ultra-Low-Bit LLM Inference on Edge Devices](https://mlsanity.com/paper/2512.06443/) (2025)
- [StableQAT: Stable Quantization-Aware Training at Ultra-Low Bitwidths](https://mlsanity.com/paper/2601.19320/) (2026)
- [D2Quant: Accurate Low-bit Post-Training Weight Quantization for LLMs](https://mlsanity.com/paper/2602.02546/) (2026)
- [FlexRank: Nested Low-Rank Knowledge Decomposition for Adaptive Model Deployment](https://mlsanity.com/paper/2602.02680/) (2026)
- [JacQuant: STE-Free Quantization-Aware Training via Learned Jacobian Surrogates](https://mlsanity.com/paper/2605.25469/) (2026)
- [UniSVQ: 2-bit Unified Scalar-Vector Quantization](https://mlsanity.com/paper/2606.10520/) (2026)
- [Resource-Efficient Language Models: Quantization for Fast and Accessible Inference](https://mlsanity.com/paper/2505.08620/) (2025)
- [Rethinking 1-bit Optimization Leveraging Pre-trained Large Language Models](https://mlsanity.com/paper/2508.06974/) (2025)
- [SFMP: Fine-Grained, Hardware-Friendly and Search-Free Mixed-Precision Quantization for Large Language Models](https://mlsanity.com/paper/2602.01027/) (2026)

## Abstract (arXiv)

The optimal bit-width for achieving the best trade-off between quantized model size and accuracy has been a subject of ongoing debate. While some advocate for 4-bit quantization, others propose that 1.58-bit offers superior results. However, the lack of a cohesive framework for different bits has left such conclusions relatively tenuous. We present ParetoQ, the first unified framework that facilitates rigorous comparisons across 1-bit, 1.58-bit, 2-bit, 3-bit, and 4-bit quantization settings. Our findings reveal a notable learning transition between 2 and 3 bits: For 3-bits and above, the fine-tuned models stay close to their original pre-trained distributions, whereas for learning 2-bit networks or below, the representations change drastically. By optimizing training schemes and refining quantization functions, ParetoQ surpasses all previous methods tailored to specific bit widths. Remarkably, our ParetoQ ternary 600M-parameter model even outperforms the previous SoTA ternary 3B-parameter model in accuracy, using only one-fifth of the parameters. Extensive experimentation shows that ternary, 2-bit, and 3-bit quantization maintains comparable performance in the size-accuracy trade-off and generally exceeds 4-bit and binary quantization. Considering hardware constraints, 2-bit quantization offers promising potential for memory reduction and speedup.