# R2-Router: A New Paradigm for LLM Routing with Reasoning

arXiv: https://arxiv.org/abs/2602.02823 · HTML: https://mlsanity.com/paper/2602.02823/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 24 of its 45 references that are in the map.
- First posted on arXiv 2 Feb 2026.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 24 of 24)

- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Token-Budget-Aware LLM Reasoning](https://mlsanity.com/paper/2412.18547/) (2024)
- [RouterBench: A Benchmark for Multi-LLM Routing System](https://mlsanity.com/paper/2403.12031/) (2024)
- [GraphRouter: A Graph-based Router for LLM Selections](https://mlsanity.com/paper/2410.03834/) (2024)
- [The Impact of Reasoning Step Length on Large Language Models](https://mlsanity.com/paper/2401.04925/) (2024)
- [AutoMix: Automatically Mixing Language Models](https://mlsanity.com/paper/2310.12963/) (2023)
- [RAGBench: Explainable Benchmark for Retrieval-Augmented Generation Systems](https://mlsanity.com/paper/2407.11005/) (2024)
- [Concise Thoughts: Impact of Output Length on LLM Reasoning and Cost](https://mlsanity.com/paper/2407.19825/) (2024)
- [How Well do LLMs Compress Their Own Chain-of-Thought? A Token Complexity Approach](https://mlsanity.com/paper/2503.01141/) (2025)
- [Universal Model Routing for Efficient LLM Inference](https://mlsanity.com/paper/2502.08773/) (2025)
- [RouterEval: A Comprehensive Benchmark for Routing LLMs to Explore Model-level Scaling Up in LLMs](https://mlsanity.com/paper/2503.10657/) (2025)
- [TensorOpera Router: A Multi-Model Router for Efficient LLM Inference](https://mlsanity.com/paper/2408.12320/) (2024)
- [MetaLLM: A High-performant and Cost-efficient Dynamic Framework for Wrapping LLMs](https://mlsanity.com/paper/2407.10834/) (2024)
- [IRT-Router: Effective and Interpretable Multi-LLM Routing via Item Response Theory](https://mlsanity.com/paper/2506.01048/) (2025)
- [CARROT: A Cost Aware Rate Optimal Router](https://mlsanity.com/paper/2502.03261/) (2025)
- [OmniRouter: Budget and Performance Controllable Multi-LLM Routing](https://mlsanity.com/paper/2502.20576/) (2025)
- [Route to Reason: Adaptive Routing for LLM and Reasoning Strategy Selection](https://mlsanity.com/paper/2505.19435/) (2025)
- [RouterArena: An Open Platform for Comprehensive Comparison of LLM Routers](https://mlsanity.com/paper/2510.00202/) (2025)
- [Cost-Aware Contrastive Routing for LLMs](https://mlsanity.com/paper/2508.12491/) (2025)
- [Route-and-Reason: Scaling Large Language Model Reasoning with Reinforced Model Router](https://mlsanity.com/paper/2506.05901/) (2025)
- [When to Reason: Semantic Router for vLLM](https://mlsanity.com/paper/2510.08731/) (2025)
- [How Robust Are Router-LLMs? Analysis of the Fragility of LLM Routing Capabilities](https://mlsanity.com/paper/2504.07113/) (2025)
- [FusionFactory: Fusing LLM Capabilities with Multi-LLM Log Data](https://mlsanity.com/paper/2507.10540/) (2025)
- [Think When Needed: Model-Aware Reasoning Routing for LLM-based Ranking](https://mlsanity.com/paper/2601.18146/) (2026)

## Built on by (most cited first, 14 of 14)

- [Unsolvability Ceiling in Multi-LLM Routing: An Empirical Study of Evaluation Artifacts](https://mlsanity.com/paper/2605.07395/) (2026)
- [Switchcraft: AI Model Router for Agentic Tool Calling](https://mlsanity.com/paper/2605.07112/) (2026)
- [Learning What to Retain: Gated-Memory Routing for Efficient Collaboration in Multi-Agent LLM Systems](https://mlsanity.com/paper/2609.00237/) (2026)
- [98× Faster LLM Routing Without a Dedicated GPU: Flash Attention, Prompt Compression, and Near-Streaming for the vLLM Semantic Router](https://mlsanity.com/paper/2603.12646/) (2026)
- [LatentRouter: Can We Choose the Right Multimodal Model Before Seeing Its Answer?](https://mlsanity.com/paper/2605.11301/) (2026)
- [The Workload-Router-Pool Architecture for LLM Inference Optimization: A Vision Paper from the vLLM Semantic Router Project](https://mlsanity.com/paper/2603.21354/) (2026)
- [The Routing Plateau: Understanding the Accuracy Limits of LLM Routers](https://mlsanity.com/paper/2606.07587/) (2026)
- [TRACE-ROUTER: Task-Consistent and Adaptive Online Routing for Agentic AI](https://mlsanity.com/paper/2607.22465/) (2026)
- [Natural Language Query to Configuration for Retrieval Agents](https://mlsanity.com/paper/2605.27361/) (2026)
- [IR3DE: A Linear Router for Large Language Models](https://mlsanity.com/paper/2606.06098/) (2026)
- [INFRAMIND: Infrastructure-Aware Multi-Agent Orchestration](https://mlsanity.com/paper/2606.11440/) (2026)
- [REALM: A Unified Red-Teaming Benchmark for Physical-World VLMs](https://mlsanity.com/paper/2606.23892/) (2026)
- [Routing Without Embeddings: Fast And Interpretable Routing With Regular Expressions](https://mlsanity.com/paper/2609.34326/) (2026)
- [Rethinking Quality in Large Language Model Routing: Security-Conditioned Model-and-Budget Selection](https://mlsanity.com/paper/2610.18566/) (2026)

## Abstract (arXiv)

As LLMs proliferate with diverse capabilities and costs, LLM routing has emerged by learning to predict each LLM's quality and cost for a given query, then selecting the one with high quality and low cost. However, existing routers implicitly assume a single fixed quality and cost per LLM for each query, ignoring that the same LLM's quality varies with its output length. This causes routers to exclude powerful LLMs when their estimated cost exceeds the budget, missing the opportunity that these LLMs could still deliver high quality at reduced cost with shorter outputs. To address this, we introduce R2-Router, which treats output length budget as a controllable variable and jointly selects the best LLM and length budget, enforcing the budget via length-constrained instructions. This enables R2-Router to discover that a powerful LLM with constrained output can outperform a weaker LLM at comparable cost-efficient configurations invisible to prior methods. Together with the router framework, we construct R2-Bench, the first routing dataset capturing LLM behavior across diverse output length budgets. Experiments show that R2-Router achieves state-of-the-art performance at 4-5× lower cost compared with existing routers. This work opens a new direction: routing as reasoning, where routers evolve from reactive selectors to deliberate reasoners that explore which LLM to use and at what cost budget. The code is publicly available at https://github.com/UCF-ML-Research/R2-Router.