# Understanding INT4 Quantization for Transformer Models: Latency Speedup, Composability, and Failure Cases

arXiv: https://arxiv.org/abs/2301.12017 · HTML: https://mlsanity.com/paper/2301.12017/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Builds on 57 of its 73 references that are in the map.
- First posted on arXiv 27 Jan 2023.

## Builds on (most cited first, 30 of 57)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://mlsanity.com/paper/1910.01108/) (2019)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Pointer Sentinel Mixture Models](https://mlsanity.com/paper/1609.07843/) (2016)
- [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://mlsanity.com/paper/1909.11942/) (2019)
- [A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference](https://mlsanity.com/paper/1704.05426/) (2017)
- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [Learning both Weights and Connections for Efficient Neural Networks](https://mlsanity.com/paper/1506.02626/) (2015)
- [GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers](https://mlsanity.com/paper/2210.17323/) (2022)
- [MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers](https://mlsanity.com/paper/2002.10957/) (2020)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization](https://mlsanity.com/paper/1808.08745/) (2018)
- [TinyBERT: Distilling BERT for Natural Language Understanding](https://mlsanity.com/paper/1909.10351/) (2019)
- [Mixed Precision Training](https://mlsanity.com/paper/1710.03740/) (2017)
- [Pruning Filters for Efficient ConvNets](https://mlsanity.com/paper/1608.08710/) (2016)
- [BERT Rediscovers the Classical NLP Pipeline](https://mlsanity.com/paper/1905.05950/) (2019)
- [Teaching Machines to Read and Comprehend](https://mlsanity.com/paper/1506.03340/) (2015)
- [SemEval-2017 Task 1: Semantic Textual Similarity - Multilingual and Cross-lingual Focused Evaluation](https://mlsanity.com/paper/1708.00055/) (2017)
- [Neural Network Acceptability Judgments](https://mlsanity.com/paper/1805.12471/) (2018)
- [On Layer Normalization in the Transformer Architecture](https://mlsanity.com/paper/2002.04745/) (2020)
- [LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale](https://mlsanity.com/paper/2208.07339/) (2022)
- [Are Sixteen Heads Really Better than One?](https://mlsanity.com/paper/1905.10650/) (2019)
- [A Survey of Quantization Methods for Efficient Neural Network Inference](https://mlsanity.com/paper/2103.13630/) (2021)
- [Universal Transformers](https://mlsanity.com/paper/1807.03819/) (2018)
- [ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers](https://mlsanity.com/paper/2206.01861/) (2022)

## Built on by (most cited first, 23 of 23)

- [Efficient Large Language Models: A Survey](https://mlsanity.com/paper/2312.03863/) (2023)
- [Active-Dormant Attention Heads: Mechanistically Demystifying Extreme-Token Phenomena in LLMs](https://mlsanity.com/paper/2410.13835/) (2024)
- [Optimizing Large Language Model Training Using FP4 Quantization](https://mlsanity.com/paper/2501.17116/) (2025)
- [Mitigating the Impact of Outlier Channels for Language Model Quantization with Activation Regularization](https://mlsanity.com/paper/2404.03605/) (2024)
- [Token-Scaled Logit Distillation for Ternary Weight Generative Language Models](https://mlsanity.com/paper/2308.06744/) (2023)
- [Vocabulary-Defined Semantics: Latent Space Clustering for Improving In-Context Learning](https://mlsanity.com/paper/2401.16184/) (2024)
- [A Semantic-Aware Layer-Freezing Approach to Computation-Efficient Fine-Tuning of Language Models](https://mlsanity.com/paper/2406.11753/) (2024)
- [A2Q+: Improving Accumulator-Aware Weight Quantization](https://mlsanity.com/paper/2401.10432/) (2024)
- [OutlierTune: Efficient Channel-Wise Quantization for Large Language Models](https://mlsanity.com/paper/2406.18832/) (2024)
- [4bit-Quantization in Vector-Embedding for RAG](https://mlsanity.com/paper/2501.10534/) (2025)
- [MergeQuant: Accurate 4-bit Static Quantization of Large Language Models by Channel-wise Calibration](https://mlsanity.com/paper/2503.07654/) (2025)
- [Dynamically Allocated Interval-Based Generative Linguistic Steganography with Roulette Wheel](https://mlsanity.com/paper/2401.15656/) (2024)
- [AERO: Entropy-Guided Framework for Private LLM Inference](https://mlsanity.com/paper/2410.13060/) (2024)
- [LegalEval-Q: A New Benchmark for The Quality Evaluation of LLM-Generated Legal Text](https://mlsanity.com/paper/2505.24826/) (2025)
- [Prune-then-Quantize or Quantize-then-Prune? Understanding the Impact of Compression Order in Joint Model Compression](https://mlsanity.com/paper/2603.18426/) (2026)
- [How to Parameterize Asymmetric Quantization Ranges for Quantization-Aware Training](https://mlsanity.com/paper/2404.16898/) (2024)
- [Low-Rank Correction for Quantized LLMs](https://mlsanity.com/paper/2412.07902/) (2024)
- [HOT: Hadamard-based Optimized Training](https://mlsanity.com/paper/2503.21261/) (2025)
- [BATON: Enhancing Batch-wise Inference Efficiency for Large Language Models via Dynamic Re-batching](https://mlsanity.com/paper/2410.18701/) (2024)
- [iServe: An Intent-based Serving System for LLMs](https://mlsanity.com/paper/2501.13111/) (2025)
- [Not All Attention is Needed: Parameter and Computation Efficient Transfer Learning for Multi-modal Large Language Models](https://mlsanity.com/paper/2403.15226/) (2024)
- [LLM-CoOpt: A Co-Design and Optimization Framework for Efficient LLM Inference on Heterogeneous Platforms](https://mlsanity.com/paper/2602.09323/) (2026)
- [Building Socio-Affective Artificial Intelligence for Interactive Multi-Agent Simulations](https://mlsanity.com/paper/2609.26927/) (2026)

## Abstract (arXiv)

Improving the deployment efficiency of transformer-based language models has been challenging given their high computation and memory cost. While INT8 quantization has recently been shown to be effective in reducing both the memory cost and latency while preserving model accuracy, it remains unclear whether we can leverage INT4 (which doubles peak hardware throughput) to achieve further latency improvement. In this study, we explore the feasibility of employing INT4 weight and activation (W4A4) quantization for language models. Our findings indicate that W4A4 quantization introduces no to negligible accuracy degradation for encoder-only and encoder-decoder models, but causes a significant accuracy drop for decoder-only models. To materialize the performance gain using W4A4, we develop a highly optimized end-to-end W4A4 encoder inference pipeline supporting different quantization strategies. Our INT4 pipeline is 8.5× faster for latency-oriented scenarios and up to 3× for throughput-oriented scenarios compared to the inference of FP16, and improves the SOTA BERT INT8 performance from FasterTransformer by up to 1.7×. We provide insights into the failure cases when applying W4A4 to decoder-only models, and further explore the compatibility of INT4 quantization with other compression methods, like pruning and layer reduction.