CalibQuant: 1-Bit KV Cache Quantization for Multimodal LLMs

Insu Han, Zeliang Zhang, Zhiyuan Wang, Yifan Zhu, Susan Liang, Jiani Liu, Haiting Lin, Mingjie Zhao, Chenliang Xu, Kun Wan, Wentian Zhao

Keywords: Vision language model, KV cache compression, quantization

Abstract: Multimodal Large Language Models (MLLMs) have demonstrated remarkable performance across diverse applications. However, their computational overhead during deployment remains a critical bottleneck. While Key-Value (KV) caching effectively trades memory for computation to enhance inference efficiency, the growing memory footprint from extensive KV caches significantly reduces throughput and restricts prolonged deployment on memory-constrained GPU devices. To address this challenge, we propose a simple yet highly effective visual quantization strategy that drastically reduces both memory and computational overhead. Specifically, we introduce an extreme 1-bit quantization scheme, complemented by novel post-scaling and calibration techniques tailored to the intrinsic patterns of KV caches, thereby ensuring high efficiency without compromising model performance. Leveraging Triton for runtime optimization, we achieve a {\bf 10x} throughput increase on InternVL models. Our method is designed to be plug-and-play, seamlessly integrating with various existing MLLMs without requiring architectural changes. Extensive experiments confirm that our approach significantly reduces memory usage while maintaining computational efficiency and preserving multimodal capabilities.

Submission Number: 37