# LLM-FP4: 4-Bit Floating-Point Quantized Transformers

arXiv: https://arxiv.org/abs/2310.16836 · HTML: https://mlsanity.com/paper/2310.16836/

## Key facts

- Cited by 69 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 30 references that are in the map.
- First posted on arXiv 25 Oct 2023.

## Builds on (most cited first, 25 of 25)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [LLM.int8(): 8-bit Matrix Multiplication for Transformers at Scale](https://mlsanity.com/paper/2208.07339/) (2022)
- [Up or Down? Adaptive Rounding for Post-Training Quantization](https://mlsanity.com/paper/2004.10568/) (2020)
- [BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstruction](https://mlsanity.com/paper/2102.05426/) (2021)
- [VATT: Transformers for Multimodal Self-Supervised Learning from Raw Video, Audio and Text](https://mlsanity.com/paper/2104.11178/) (2021)
- [Data-Free Quantization Through Weight Equalization and Bias Correction](https://mlsanity.com/paper/1906.04721/) (2019)
- [Post-Training Quantization for Vision Transformer](https://mlsanity.com/paper/2106.14156/) (2021)
- [ZeroQ: A Novel Zero Shot Quantization Framework](https://mlsanity.com/paper/2001.00281/) (2020)
- [Low-bit Quantization of Neural Networks for Efficient Inference](https://mlsanity.com/paper/1902.06822/) (2019)
- [QDrop: Randomly Dropping Quantization for Extremely Low-bit Post-Training Quantization](https://mlsanity.com/paper/2203.05740/) (2022)
- [PTQ4ViT: Post-training quantization for vision transformers with twin uniform quantization](https://mlsanity.com/paper/2111.12293/) (2021)
- [Understanding and Overcoming the Challenges of Efficient Transformer Quantization](https://mlsanity.com/paper/2109.12948/) (2021)
- [FP8 Quantization: The Power of the Exponent](https://mlsanity.com/paper/2208.09225/) (2022)
- [Towards Accurate Post-Training Quantization for Vision Transformer](https://mlsanity.com/paper/2303.14341/) (2023)
- [EasyQuant: Post-training Quantization via Scale Optimization](https://mlsanity.com/paper/2006.16669/) (2020)
- [Q-HyViT: Post-Training Quantization of Hybrid Vision Transformers with Bridge Block Reconstruction for IoT Systems](https://mlsanity.com/paper/2303.12557/) (2023)

## Built on by (most cited first, 30 of 69)

- [SpinQuant: LLM quantization with learned rotations](https://mlsanity.com/paper/2405.16406/) (2024)
- [Atom: Low-bit Quantization for Efficient and Accurate LLM Serving](https://mlsanity.com/paper/2310.19102/) (2023)
- [MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases](https://mlsanity.com/paper/2402.14905/) (2024)
- [A Survey on Efficient Inference for Large Language Models](https://mlsanity.com/paper/2404.14294/) (2024)
- [BitDistiller: Unleashing the Potential of Sub-4-Bit LLMs via Self-Distillation](https://mlsanity.com/paper/2402.10631/) (2024)
- [Optimizing Large Language Model Training Using FP4 Quantization](https://mlsanity.com/paper/2501.17116/) (2025)
- [BitVLA: 1-bit Vision-Language-Action Models for Robotics Manipulation](https://mlsanity.com/paper/2506.07530/) (2025)
- [Pruner-Zero: Evolving Symbolic Pruning Metric from scratch for Large Language Models](https://mlsanity.com/paper/2406.02924/) (2024)
- [Large Language Model Inference Acceleration: A Comprehensive Hardware Perspective](https://mlsanity.com/paper/2410.04466/) (2024)
- [Four Over Six: More Accurate NVFP4 Quantization with Adaptive Block Scaling](https://mlsanity.com/paper/2512.02010/) (2025)
- [APTQ: Attention-aware Post-Training Mixed-Precision Quantization for Large Language Models](https://mlsanity.com/paper/2402.14866/) (2024)
- [A Survey of Small Language Models](https://mlsanity.com/paper/2410.20011/) (2024)
- [LLaMAntino: LLaMA 2 Models for Effective Text Generation in Italian Language](https://mlsanity.com/paper/2312.09993/) (2023)
- [A Survey: Collaborative Hardware and Software Design in the Era of Large Language Models](https://mlsanity.com/paper/2410.07265/) (2024)
- [Quantization-Aware Distillation for NVFP4 Inference Accuracy Recovery](https://mlsanity.com/paper/2601.20088/) (2026)
- [Jellyfish: A Large Language Model for Data Preprocessing](https://mlsanity.com/paper/2312.01678/) (2023)
- [Transformer-Lite: High-efficiency Deployment of Large Language Models on Mobile Phone GPUs](https://mlsanity.com/paper/2403.20041/) (2024)
- [HQ-DiT: Efficient Diffusion Transformer with FP4 Hybrid Quantization](https://mlsanity.com/paper/2405.19751/) (2024)
- [ZeroQuant(4+2): Redefining LLMs Quantization with a New FP6-Centric Strategy for Diverse Generative Tasks](https://mlsanity.com/paper/2312.08583/) (2023)
- [Toward Adaptive Large Language Models Structured Pruning via Hybrid-grained Weight Importance Assessment](https://mlsanity.com/paper/2403.10799/) (2024)
- [Model Quantization and Hardware Acceleration for Vision Transformers: A Comprehensive Survey](https://mlsanity.com/paper/2405.00314/) (2024)
- [RoLoRA: Fine-tuning Rotated Outlier-free LLMs for Effective Weight-Activation Quantization](https://mlsanity.com/paper/2407.08044/) (2024)
- [SLED: A Speculative LLM Decoding Framework for Efficient Edge Serving](https://mlsanity.com/paper/2506.09397/) (2025)
- [BlockDialect: Block-wise Fine-grained Mixed Format Quantization for Energy-Efficient LLM Inference](https://mlsanity.com/paper/2501.01144/) (2025)
- [QuZO: Quantized Zeroth-Order Fine-Tuning for Large Language Models](https://mlsanity.com/paper/2502.12346/) (2025)
- [MX+: Pushing the Limits of Microscaling Formats for Efficient Large Language Model Serving](https://mlsanity.com/paper/2510.14557/) (2025)
- [A Speed Odyssey for Deployable Quantization of LLMs](https://mlsanity.com/paper/2311.09550/) (2023)
- [LUT-DLA: Lookup Table as Efficient Extreme Low-Bit Deep Learning Accelerator](https://mlsanity.com/paper/2501.10658/) (2025)
- [PatentGPT: A Large Language Model for Intellectual Property](https://mlsanity.com/paper/2404.18255/) (2024)
- [eXmY: A Data Type and Technique for Arbitrary Bit Precision Quantization](https://mlsanity.com/paper/2405.13938/) (2024)

## Abstract (arXiv)

We propose LLM-FP4 for quantizing both weights and activations in large language models (LLMs) down to 4-bit floating-point values, in a post-training manner. Existing post-training quantization (PTQ) solutions are primarily integer-based and struggle with bit widths below 8 bits. Compared to integer quantization, floating-point (FP) quantization is more flexible and can better handle long-tail or bell-shaped distributions, and it has emerged as a default choice in many hardware platforms. One characteristic of FP quantization is that its performance largely depends on the choice of exponent bits and clipping range. In this regard, we construct a strong FP-PTQ baseline by searching for the optimal quantization parameters. Furthermore, we observe a high inter-channel variance and low intra-channel variance pattern in activation distributions, which adds activation quantization difficulty. We recognize this pattern to be consistent across a spectrum of transformer models designed for diverse tasks, such as LLMs, BERT, and Vision Transformer models. To tackle this, we propose per-channel activation quantization and show that these additional scaling factors can be reparameterized as exponential biases of weights, incurring a negligible cost. Our method, for the first time, can quantize both weights and activations in the LLaMA-13B to only 4-bit and achieves an average score of 63.1 on the common sense zero-shot reasoning tasks, which is only 5.8 lower than the full-precision model, significantly outperforming the previous state-of-the-art by 12.7 points. Code is available at: https://github.com/nbasyl/LLM-FP4.