@article{chimeralatencyandperformanceawaremultiag, title = {Chimera: Latency- and Performance-Aware Multi-agent Serving for Heterogeneous LLMs}, author = {Kangqi Ni and Wenyue Hua and Xiaoxiang Shi and Jiang Guo and Shiyu Chang and Tianlong Chen}, year = {2026}, eprint = {2603.22206}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2603.22206}, }