Paper Detail
Tao Feng, Fangxu Yu, Haozhen Zhang, Zhongjie Dai, Liangqi Yuan, Zijie Lei, Weizhi Zhang, Kunlun Zhu, Haodong Yue, Keyang Xuan, Ge Liu, Jiaxuan You
No single large language model (LLM) is optimal across all queries and budget constraints, making model routing essential for cost-effective deployment. Existing routers adopt diverse formulations and implementations, making fair comparison and extension difficult. We present a unified formulation of LLM routing as a sequential decision process characterized by five components: context encoders, model encoders, scoring functions, decision rules, and learning signals, covering single-turn, multi-turn, and personalized routing. Based on this formulation, we develop an automated pipeline for constructing routing supervision and evaluating routers jointly on response quality and inference cost. The resulting benchmark, xRouteBench, spans generic LLM, memory-augmented, vision, time-series, and personalized routing tasks. We further introduce LLMRouter, an open-source modular infrastructure with more than 16 representative routers. Our empirical study shows that learned routers outperform the strongest fixed-model baseline by 14.6% relatively, lightweight routers become more competitive under tight cost constraints, and user-conditioned routing consistently improves personalization.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@misc{feng2026llmrouter,
title = {LLMRouter: Unified Infrastructure for Developing, Evaluating, and Deploying LLM Routers},
author = {Tao Feng and Fangxu Yu and Haozhen Zhang and Zhongjie Dai and Liangqi Yuan and Zijie Lei and Weizhi Zhang and Kunlun Zhu and Haodong Yue and Keyang Xuan and Ge Liu and Jiaxuan You},
year = {2026},
abstract = {No single large language model (LLM) is optimal across all queries and budget constraints, making model routing essential for cost-effective deployment. Existing routers adopt diverse formulations and implementations, making fair comparison and extension difficult. We present a unified formulation of LLM routing as a sequential decision process characterized by five components: context encoders, model encoders, scoring functions, decision rules, and learning signals, covering single-turn, multi-},
url = {https://huggingface.co/papers/2608.06867},
keywords = {LLM routing, context encoders, model encoders, scoring functions, decision rules, learning signals, multi-turn routing, personalized routing, xRouteBench, LLMRouter, learned routers, user-conditioned routing, code available, huggingface daily},
eprint = {2608.06867},
archiveprefix = {arXiv},
}
{}