@article{prismunleashinggpusharingforcost, title = {Prism: Unleashing GPU Sharing for Cost-Efficient Multi-LLM Serving}, author = {Shan Yu and Jiarong Xing and Yifan Qiao and Mingyuan Ma and Yangmin Li and Yang Wang and Shuo Yang and Zhiqiang Xie and Shiyi Cao and Ke Bao and Ion Stoica and Harry Xu and Ying Sheng}, year = {2025}, eprint = {2505.04021}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2505.04021v2}, }