xRouter: Training Cost-Aware LLMs Orchestration System via Reinforcement Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911200304955392 |
|---|---|
| author | Qian, Cheng Liu, Zuxin Kokane, Shirley Prabhakar, Akshara Qiu, Jielin Chen, Haolin Liu, Zhiwei Ji, Heng Yao, Weiran Heinecke, Shelby Savarese, Silvio Xiong, Caiming Wang, Huan |
| author_facet | Qian, Cheng Liu, Zuxin Kokane, Shirley Prabhakar, Akshara Qiu, Jielin Chen, Haolin Liu, Zhiwei Ji, Heng Yao, Weiran Heinecke, Shelby Savarese, Silvio Xiong, Caiming Wang, Huan |
| contents | Modern LLM deployments confront a widening cost-performance spectrum: premium models deliver strong reasoning but are expensive, while lightweight models are economical yet brittle on complex tasks. Static escalation rules and keyword heuristics under-utilize this spectrum and fail to adapt across task types. We present xRouter, a tool-calling-based routing system in which a learned router can either answer directly or invoke one or more external models. The router is trained end-to-end with reinforcement learning using an explicit, cost-aware reward that encodes cost-performance trade-offs, eliminating the need for hand-engineered routing rules. Our implementation encompasses the full reinforcement learning framework, including reward and cost accounting, as well as the deployment and evaluation pipelines. Across diverse benchmarks, xRouter achieves strong cost-performance trade-offs (e.g., substantial cost reductions at comparable task completion rates), and provides empirical insights into what reliably helps learned routing and what does not, ranging from model trainability to the difficulty of eliciting sophisticated orchestration behaviors in small open models. We hope these findings and our open implementation will serve as a practical substrate for advancing learned, cost-aware LLM orchestration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_08439 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | xRouter: Training Cost-Aware LLMs Orchestration System via Reinforcement Learning Qian, Cheng Liu, Zuxin Kokane, Shirley Prabhakar, Akshara Qiu, Jielin Chen, Haolin Liu, Zhiwei Ji, Heng Yao, Weiran Heinecke, Shelby Savarese, Silvio Xiong, Caiming Wang, Huan Machine Learning Artificial Intelligence Computation and Language Modern LLM deployments confront a widening cost-performance spectrum: premium models deliver strong reasoning but are expensive, while lightweight models are economical yet brittle on complex tasks. Static escalation rules and keyword heuristics under-utilize this spectrum and fail to adapt across task types. We present xRouter, a tool-calling-based routing system in which a learned router can either answer directly or invoke one or more external models. The router is trained end-to-end with reinforcement learning using an explicit, cost-aware reward that encodes cost-performance trade-offs, eliminating the need for hand-engineered routing rules. Our implementation encompasses the full reinforcement learning framework, including reward and cost accounting, as well as the deployment and evaluation pipelines. Across diverse benchmarks, xRouter achieves strong cost-performance trade-offs (e.g., substantial cost reductions at comparable task completion rates), and provides empirical insights into what reliably helps learned routing and what does not, ranging from model trainability to the difficulty of eliciting sophisticated orchestration behaviors in small open models. We hope these findings and our open implementation will serve as a practical substrate for advancing learned, cost-aware LLM orchestration. |
| title | xRouter: Training Cost-Aware LLMs Orchestration System via Reinforcement Learning |
| topic | Machine Learning Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2510.08439 |