The growing computational demands of training large language models (LLMs) necessitate more efficient methods. Quantized training presents a promising solution by enabling low-bit arithmetic operations to reduce these costs. While FP8 precision has demonstrated feasibility, leveraging FP4 remains a challenge due to significant quantization errors and limited representational capacity. This work introduces the first FP4 training framework for LLMs, addressing these challenges with two key innovations: a differentiable quantization estimator for precise weight updates and an outlier clamping and compensation strategy to prevent activation collapse. To ensure stability, the framework integrates a mixed-precision training scheme and vector-wise quantization. Experimental results demonstrate that our FP4 framework achieves accuracy comparable to BF16 and FP8, with minimal degradation, scaling effectively to 13B-parameter LLMs trained on up to 100B tokens. With the emergence of next-generation hardware supporting FP4, our framework sets a foundation for efficient ultra-low precision training.
@article{arxiv.2501.17116,
title = {Optimizing Large Language Model Training Using FP4 Quantization},
author = {Ruizhe Wang and Yeyun Gong and Xiao Liu and Guoshuai Zhao and Ziyue Yang and Baining Guo and Zhengjun Zha and Peng Cheng},
journal= {arXiv preprint arXiv:2501.17116},
year = {2026}
}