SVDq: 1.25-bit and 410x Key Cache Compression for LLM Attention

Hong Yankun, Li Xing, Zhen Hui-Ling, Yu Xianzhi, Liu Wulong, Yuan Mingxuan
January 2025
Cite
Type
Journal article
Publication
arXiv preprint arXiv:2502.15304

Add the full text or supplementary notes for the publication here using Markdown formatting.