mlsanityData to 7 Oct 2026

MasRouter: Learning to Route LLMs for Multi-Agent Systems

arXiv 2502.11133 ↗First posted 16 Feb 2025cs.LG cs.MAcited by 80

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

Multi-agent systems (MAS) powered by Large Language Models (LLMs) have been demonstrated to push the boundaries of LLM capabilities, yet they often incur significant costs and face challenges in dynamic LLM selection. Current LLM routing methods effectively reduce overhead in single-agent scenarios by customizing LLM selection for each query, but they overlook the critical decisions regarding collaboration modes and agent roles in MAS. In response to this challenge, we first introduce the problem of Multi-Agent System Routing (MASR), which integrates all components of MAS into a unified routing framework. Toward this goal, we propose MasRouter, the first high-performing, cost-effective, and inductive MASR solution. MasRouter employs collaboration mode determination, role allocation, and LLM routing through a cascaded controller network, progressively constructing a MAS that balances effectiveness and efficiency. Extensive experiments demonstrate that MasRouter is (1) high-performing, achieving a 1.8%~8.2% improvement over the state-of-the-art method on MBPP; (2) economical, reducing overhead by up to 52.07% compared to SOTA methods on HumanEval; and (3) plug-and-play, seamlessly integrating with mainstream MAS frameworks, reducing overhead by 17.21%~28.17% via customized routing. The code is available at https://github.com/yanweiyue/masrouter.

Key facts. Cited by 80 papers in the map (data through 7 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 42 of its 78 references that are in the map. First posted on arXiv 16 Feb 2025.

Builds on

42 of 78 references.

Closest ancestors · 5

Same direction, newest first

  1. 2024AgentSquare: Automatic LLM Agent Search in Modular Design Space8 Oct
  2. 2024On the Resilience of LLM-Based Multi-Agent Collaboration with Faulty Agents2 Aug
  3. 2024Small LLMs Are Weak Tool Learners: A Multi-LLM Agent14 Jan
  4. 2023AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors21 Aug
  5. 2023Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration11 Jul

Other references · 37

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2024The Llama 3 Herd of Modelscited by 14,192
  3. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  4. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
  5. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  6. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  7. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  8. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
  9. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  10. 2021Program Synthesis with Large Language Modelscited by 3,509
  11. 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,175
  12. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,739
25 more
  1. 2020MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformerscited by 1,414
  2. 2024DeepSeek-Coder: When the Large Language Model Meets Programming -- The Rise of Code Intelligencecited by 1,154
  3. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debatecited by 995
  4. 2023FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performancecited by 791
  5. 2024RouteLLM: Learning to Route LLMs with Preference Datacited by 633
  6. 2022Complexity-Based Prompting for Multi-Step Reasoningcited by 387
  7. 2024Hybrid LLM: Cost-Efficient and Quality-Aware Query Routingcited by 354
  8. 2023Exploring Collaboration Mechanisms for LLM Agents: A Social Psychology Viewcited by 221
  9. 2024GraphRouter: A Graph-based Router for LLM Selectionscited by 154
  10. 2023Fly-Swat or Cannon? Cost-Effective Language Model Choice via Meta-Modelingcited by 94
  11. 2024Self-Organized Agents: A LLM Multi-Agent Framework toward Ultra Large-Scale Code Generation and Optimizationcited by 84
  12. 2023Exchange-of-Thought: Enhancing Large Language Model Capabilities through Cross-Model Communicationcited by 78
  13. 2024Harder Tasks Need More Experts: Dynamic Routing in MoE Modelscited by 71
  14. 2024TensorOpera Router: A Multi-Model Router for Efficient LLM Inferencecited by 59
  15. 2024Harnessing the Power of Multiple Minds: Lessons Learned from LLM Routingcited by 43
  16. 2024Cost-Effective Online Multi-LLM Selection with Versatile Reward Modelscited by 36
  17. 2024Routoo: Learning to Route to Large Language Models Effectivelycited by 34
  18. 2025LLM-based Human Simulations Have Not Yet Been Reliablecited by 30
  19. 2024TasTe: Teaching Large Language Models to Translate through Self-Reflectioncited by 27
  20. 2024Small Language Models are Good Too: An Empirical Study of Zero-Shot Classificationcited by 17
  21. 2023Math Agents: Computational Infrastructure, Mathematical Embedding, and Genomicscited by 15
  22. 2024DA-MoE: Towards Dynamic Expert Allocation for Mixture-of-Experts Modelscited by 15
  23. 2024Automating Customer Needs Analysis: A Comparative Study of Large Language Models in the Travel Industrycited by 6
  24. 2020More than Code: Contributions in Scrum Software Engineering Teamscited by 1
  25. 2023Detecting and Optimising Team Interactions in Software Developmentcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 25; 2026: 55.

80 papers.

Sort:

By area:

  1. 2025G-Memory: Tracing Hierarchical Memory for Multi-Agent Systemscited by 112
  2. 2025A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deploymentcited by 110
  3. 2025AgenTracer: Who Is Inducing Failure in the LLM Agentic Systems?cited by 105
  4. 2025ReMA: Learning to Meta-think for LLMs with Multi-Agent Reinforcement Learningcited by 78
  5. 2026Natural-Language Agent Harnessescited by 54
  6. 2026A Survey of Agentic Reasoning for Large Language Models: Towards Recursively Self-Improving and Collective Agentscited by 47
  7. 2025Don't Overthink It: A Survey of Efficient R1-style Large Reasoning Modelscited by 36
  8. 2026When Single-Agent with Skills Replace Multi-Agent Systems and When They Failcited by 34
  9. 2025X-MAS: Towards Building Multi-Agent Systems with Heterogeneous LLMscited by 33
  10. 2025Difficulty-Aware Agentic Orchestration for Query-Specific Multi-Agent Workflowscited by 23
  11. 2025Learning to Orchestrate Agents in Natural Language with the Conductorcited by 22
  12. 2026Rethinking the Value of Multi-Agent Workflow: A Strong Single Agent Baselinecited by 18
38 more
  1. 2025Maestro: Learning to Collaborate via Conditional Listwise Policy Optimization for Multi-Agent LLMscited by 14
  2. 2026EvolveRouter: Co-Evolving Routing and Prompt for Multi-Agent Question Answeringcited by 13
  3. 2026LatentMem: Customizing Latent Memory for Multi-Agent Systemscited by 12
  4. 2026Agent-as-a-Router: Agentic Model Routing for Coding Taskscited by 10
  5. 2026Exploring Knowledge Purification in Multi-Teacher Knowledge Distillation for LLMscited by 9
  6. 2026Adaptive Collaboration with Humans: Metacognitive Policy Optimization for Multi-Agent LLMs with Continual Learningcited by 9
  7. 2026FALAT: Tracing Failures in LLM Agent Trajectories via Dependency-Guided Searchcited by 9
  8. 2025LatentEvolve: Self-Evolving Test-Time Scaling in Latent Spacecited by 8
  9. 2025S-DAG: A Subject-Based Directed Acyclic Graph for Multi-Agent Heterogeneous Reasoningcited by 8
  10. 2026MetaGen: Self-Evolving Roles and Topologies for Multi-Agent LLM Reasoningcited by 8
  11. 2025DualTAP: A Dual-Task Adversarial Protector for Mobile MLLM Agentscited by 7
  12. 2026Step-level Optimization for Efficient Computer-use Agentscited by 7
  13. 2026FlowSteer: Prompt-Only Workflow Steering Exposes Planning-Time Vulnerabilities in Multi-Agent LLM Systemscited by 7
  14. 2025Federation of Agents: A Semantics-Aware Communication Fabric for Large-Scale Agentic AIcited by 6
  15. 2025Agentic Services Computingcited by 6
  16. 2025Parallelized Planning-Acting for Efficient LLM-based Multi-Agent Systems in Minecraftcited by 5
  17. 2025LAMeTA: Intent-Aware Agentic Network Optimization via a Large AI Model-Empowered Two-Stage Approachcited by 5
  18. 2025Topological Structure Learning Should Be A Research Priority for LLM-Based Multi-Agent Systemscited by 5
  19. 2025Adaptive Minds: Empowering Agents with LoRA-as-Toolscited by 5
  20. 2025TRINITY: An Evolved LLM Coordinatorcited by 5
  21. 2025How can we assess human-agent interactions? Case studies in software agent designcited by 4
  22. 2026Sakana Fugu Technical Reportcited by 4
  23. 2026Agentic Routing: The Harness-Native Data Flywheelcited by 4
  24. 2026GenRouter: Unified Workflow Routing for Agentic Image Generationcited by 4
  25. 2026Explainable Model Routing for Agentic Workflowscited by 3
  26. 2026Gated Coordination for Efficient Multi-Agent Collaboration in Minecraft Gamecited by 3
  27. 2026Learning Agent Routing From Early Experiencecited by 3
  28. 2026Optimizing the Cost-Quality Tradeoff of Agentic Theorem Provers in Leancited by 3
  29. 2025MasHost Builds It All: Autonomous Multi-Agent System Directed by Reinforcement Learningcited by 2
  30. 2025Toward Real-World Table Agents: Capabilities, Workflows, and Design Principles for LLM-based Table Intelligencecited by 2
  31. 2025textttReMind: Understanding Deductive Code Reasoning in LLMscited by 2
  32. 2026Efficient and Interpretable Multi-Agent LLM Routing via Ant Colony Optimizationcited by 2
  33. 2026EMS: Multi-Agent Voting via Efficient Majority-then-Stoppingcited by 2
  34. 2026Uno-Orchestra: Parsimonious Agent Routing via Selective Delegationcited by 2
  35. 2026Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systemscited by 2
  36. 2026FlowBank: Query-Adaptive Agentic Workflows Optimization through Precompute-and-Reusecited by 2
  37. 2025MAS2: Self-Generative, Self-Configuring, Self-Rectifying Multi-Agent Systemscited by 1
  38. 2025The Hunger Game Debate: On the Emergence of Over-Competition in Multi-Agent Systemscited by 1

50 most cited of 80 All 80 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026SC-MAS: Constructing Cost-Efficient Multi-Agent Systems with Edge-Level Heterogeneous Collaboration
  2. 2025MASLab: A Unified and Comprehensive Codebase for LLM-based Multi-Agent Systems
  3. 2025RCR-Router: Efficient Role-Aware Context Routing for Multi-Agent LLM Systems with Structured Memory
  4. 2025Optimal-Agent-Selection: State-Aware Routing Framework for Efficient Multi-Agent Collaboration
  5. 2025MAS-GPT: Training LLMs to Build LLM-based Multi-Agent Systems
  6. 2026LLM-Enabled Multi-Agent Systems: Empirical Evaluation and Insights into Emerging Design Patterns & Paradigms
  7. 2024A Unified Approach to Routing and Cascading for LLMs
  8. 2026RSI-Router: Evolving Subtask-Level LLM Routing and Skills for Cost-Efficient Agents