FlashSVD v1.5: Making Low-Rank Transformers Inference Actually Fast

作者：Wenhao Wu, Zishan Shao, Kangning Cui, Jinhee Kim, Yixiao Wang, Hancheng Ye, Danyang Zhuo, Yiran Chen · 发表于：arXiv (Cornell University) · 年份：2026 · DOI：10.48550/arxiv.2605.08314 · 研究领域：Parallel Computing and Optimization Techniques、Advanced Data Storage Technologies、Algorithms and Data Compression

SVD-based Low-rank compression reduces transformer parameters and nominal FLOPs, but these savings often translate poorly into real LLM serving speedups. We show that this gap is largely a runtime problem: factorized checkpoints fragment execution paths, and the resulting overhead differs substantially between prefill and autoregressive decode. We present FlashSVD v1.5, a unified inference runtime for serving SVD-compressed transformers. FlashSVD v1.5 maps diverse public SVD compression families to a common factorized representation and combines phase-specific kernels with dense-KV decode, packed MLP execution, and per-layer CUDA-graph replay to reorganize the low-rank serving path into a thin runtime. Across representative decoder-serving settings, FlashSVD v1.5 achieves up to 2.55x decode and 2.39x end-to-end speedup, and it attains 1.48x average decode and 1.44x average end-to-end speedup across multiple popular SVD compression families. These results suggest that practical low-rank acceleration requires runtime co-design, not compression algorithms alone. Our code is available at: https://github.com/Zishan-Shao/FlashSVD.