This is an awesome list of the best multi-agent research papers, compiled by the Swarms Team. Our mission at Swarms is to research multi-agent systems. Join our Discord Now!

---

Format

• [Paper Name] [Description] [Link]

---

Multi-Agent Collaboration & System Design

• Thought Communication in Multiagent Collaboration
• MAS-Zero: Designing Multi-Agent Systems with Zero Supervision
• K-Level Reasoning with Large Language Models
• More Agents is All You Need
• LongAgent: Scaling Language Models to 128k Context through Multi-Agent Collaboration
• AgentScope: A Flexible yet Robust Multi-Agent Platform
• Learning to Decode Collaboratively with Multiple Language Models
• AIOS: LLM Agent Operating System
• AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation
• Chain of Agents: Large Language Models Collaborating on Long-Context Tasks
• Mixture-of-Agents Enhances Large Language Model Capabilities
• EvoAgent: Towards Automatic Multi-Agent Generation via Evolutionary Algorithms
• Internet of Agents: Weaving a Web of Heterogeneous Agents for Collaborative Intelligence
• Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoning
• Optimizing Collaboration of LLM based Agents
• LLM-Agent-UMF: LLM-based Agent Unified Modeling Framework
• Optima: Optimizing Effectiveness and Efficiency for LLM-Based Multi-Agent System
• SwarmAgentic: Towards Fully Automated Agentic System Generation via Swarm Intelligence
• AGENTSNET: Coordination and Collaborative Reasoning in Multi-Agent LLMs
• LLM Economist: Large Population Models and Mechanism Design in Multi-Agent Generative Simulacra
• SOTOPIA-RL: REWARD DESIGN FOR SOCIAL INTELLIGENCE
• Talk Isn't Always Cheap: Understanding Failure Modes in Multi-Agent Debate
• Virtual Agent Economies
• Federation of Agents: A Semantics-Aware Communication Fabric for Large-Scale Agentic AI
• In-the-Flow Agentic System Optimization for Effective Planning and Tool Use
• Rethinking Mixture-of-Agents: Is Mixing Different Large Language Models Beneficial?
• TUMIX: Multi-Agent Test-Time Scaling with Tool-Use Mixture
• Achilles Heel of Distributed Multi-Agent Systems
• CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society
• Multi-Agent Collaboration: Harnessing the Power of Intelligent LLM Agents
• Rethinking the Bounds of LLM Reasoning: Are Multi-Agent Discussions the Key?
• MAgIC: Investigation of Large Language Model Powered Multi-Agent in Cognition, Adaptability, Rationality and Collaboration
• LLM Harmony: Multi-Agent Communication for Problem Solving
• Multi-Agent Consensus Seeking via Large Language Models
• Shall We Team Up: Exploring Spontaneous Cooperation of Competing LLM Agents
• Cooperative Strategic Planning Enhances Reasoning Capabilities in Large Language Models
• SMoA: Improving Multi-agent Large Language Models with Sparse Mixture-of-Agents
• Self-Adaptive Large Language Model (LLM)-Based Multiagent Systems
• A Strategic Coordination Framework of Small LLMs Matches Large LLMs in Data Synthesis
• Cross-environment Cooperation Enables Zero-shot Multi-agent Coordination
• Unlocking the Power of Multi-Agent LLM for Reasoning: From Lazy Agents to Deliberation
• The Station: An Open-World Environment for AI-Driven Discovery
• Towards a Science of Scaling Agent Systems
• RouteMoA: Dynamic Routing without Pre-Inference Boosts Efficient Mixture-of-Agents
• Intelligent AI Delegation
• Does Socialization Emerge in AI Agent Society? A Case Study of Moltbook
• Multi-agent cooperation through in-context co-player inference
• SkillOrchestra: Learning to Route Agents via Skill Transfer
• Contextual Counterfactual Credit Assignment for Multi-Agent Reinforcement Learning in LLM Collaboration GitHub
• CORAL: Towards Autonomous Multi-Agent Evolution for Open-Ended Discovery
• TRINITY: An Evolved LLM Coordinator
• Learning to Orchestrate Agents in Natural Language with the Conductor
• From Skills to Talent: Organising Heterogeneous Agents as a Real-World Company
• Can AI Agents Agree?
• ARIS: Autonomous Research via Adversarial Multi-Agent Collaboration
• Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems
• AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective Reasoning
• DarkForest: Less Talk, Higher Accuracy for Multi-Agent LLMs
• AgensFlow: A Coordination-Policy Substrate for Multi-Agent Systems
• When Cloud Agents Meet Device Agents: Lessons from Hybrid Multi-Agent Systems
• Streaming Communication in Multi-Agent Reasoning
• What Should Agents Say? Action-state Communication for Efficient Multi-Agent Systems
• SearchSwarm: Towards Delegation Intelligence in Agentic LLMs for Long-Horizon Deep Research
• Decentralized Multi-Agent Systems with Shared Context
• Sakana Fugu Technical Report
• A Technical Taxonomy of LLM Agent Communication Protocols
• Will Scaling Improve Social Simulation with LLMs?
• Agensh: Scaling Organizational Intelligence to 1,024 Agents
• Self-Organizing Agent Teams Learn to Reason Together

Multi-Agent Frameworks & Benchmarks

• MALLM: Multi-Agent Large Language Models Framework
• AgentGym: Evolving Large Language Model-based Agents across Diverse Environments
• Very Large-Scale Multi-Agent Simulation in AgentScope
• AgentClinic: A Multimodal Agent Benchmark for AI in Clinical Environments
• MultiAgentBench: Evaluating the Collaboration and Competition of LLM Agents
• TheAgentCompany: Benchmarking LLM Agents on Consequential Real World Tasks
• BoxingGym: Benchmarking Progress in Automated Experimental Design
• Symphony: A Decentralized Multi-Agent Framework for Scalable Collective Intelligence
• CoMAS: Co-Evolving Multi-Agent Systems via Interaction Rewards
• The Collaboration Gap
• API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs
• AutoAgents: A Framework for Automatic Agent Generation
• MAS-GPT: Training LLMs to Build LLM-based Multi-Agent Systems
• Multi-Agent Collaboration via Evolving Orchestration
• TCAndon-Router: Adaptive Reasoning Router for Multi-Agent Collaboration
• Context Learning for Multi-Agent Discussion
• The Vision Wormhole: Latent-Space Communication in Heterogeneous Multi-Agent Systems
• EvoSkill: Automated Skill Discovery for Multi-Agent Systems

Application-Specific Multi-Agent Systems

Software Engineering

• Automated Unit Test Improvement using Large Language Models
• Experiential Co-Learning of Software-Developing Agents
• ChatDev: Communicative Agents for Software Development
• MAGIS: LLM-Based Multi-Agent Framework for GitHub Issue Resolution
• CodeR: Issue Resolving with Multi-Agent and Task Graphs
• From LLMs to LLM-based Agents for Software Engineering: A Survey
• CodexGraph: Bridging Large Language Models and Code Repositories via Code Graph Databases
• Diversity Empowers Intelligence: Integrating Expertise of Software Engineering Agents
• Large Language Model-Based Agents for Software Engineering: A Survey
• AutoSafeCoder: A Multi-Agent Framework for Securing LLM Code Generation
• RGD: Multi-LLM Based Agent Debugger via Refinement and Generation Guidance
• CoAct-1: Computer-using Agents with Coding as Actions
• Self-Organized Agents: A LLM Multi-Agent Framework toward Ultra Large-Scale Code Generation and Optimization
• Human-In-the-Loop Software Development Agents
• LLM-Based Multi-Agent Systems for Software Engineering: Vision and the Road Ahead
• LLM-Based Multi-Agent Systems for Software Engineering: Literature Review, Vision and the Road Ahead
• Tailoring with Targeted Precision: Edit-Based Agents for Open-Domain Procedure Customization
• Agents4PLC: Automating Closed-loop PLC Code Generation and Verification in Industrial Control Systems using LLM-based Agents
• Scaling long-running autonomous coding A blog by Cursor
• Towards self-driving codebases By Cursor
• Harness Design for Long-Running Application Development

Healthcare & Medical

• Agent Hospital: A Simulacrum of Hospital with Evolvable Medical Agents
• MEDCO: Medical Education Copilots Based on A Multi-Agent Framework
• Multi Agent based Medical Assistant for Edge Devices
• Can AI Agents Design and Implement Drug Discovery Pipelines?
• Sequential Diagnosis with Language Models ||Open-Source Implementation Code Link
• Towards an AI co-scientist --- Implementation
• The Virtual Lab of AI agents designs new SARS-CoV-2 nanobodies
• Cerebra: A Multidisciplinary AI Board for Multimodal Dementia Characterization and Risk Assessment
• Evolving Diagnostic Agents in a Virtual Clinical Environment
• MDAgents: An Adaptive Collaboration of LLMs for Medical Decision-Making
• Zodiac: A Cardiologist-Level LLM Framework for Multi-Agent Diagnostics
• MADD: Multi-Agent Drug Discovery Orchestra
• SciAgent: A Unified Multi-Agent System for Generalistic Scientific Reasoning

Data & ML

• LAMBDA: A Large Model Based Data Agent
• Agentic Retrieval-Augmented Generation for Time Series Analysis
• Multi-Agent Collaborative Data Selection for Efficient LLM Pretraining
• AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoML
• AutoKaggle: A Multi-Agent Framework for Autonomous Data Science Competitions
• DataLab: A Unified Platform for LLM-Powered Business Intelligence

Security

• BreachSeek: A Multi-Agent Automated Penetration Tester
• Got a Secret? LLM Agents Can't Keep It: Evaluating Privacy in Multi-Agent Systems
• Mind Viruses: Self-Propagating Ideas in Multi-Agent LLM Systems Link

Multimodal

• Mora: Enabling Generalist Video Generation via A Multi-Agent Framework
• Mobile-Agent-v2: Mobile Device Operation Assistant with Effective Navigation
• Optimus-1: Hybrid Multimodal Memory Empowered Agents Excel in Long-Horizon Tasks
• Soap2Soap: Long Cinematic Video Remaking via Multi-Agent Collaboration
• Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players
• Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation
• MP5: A Multi-modal Open-ended Embodied System in Minecraft via Active Perception
• PC-Agent: A Hierarchical Multi-Agent Framework for Complex Task Automation on PC
• Improving Large Vision and Language Models by Learning from a Panel of Peers
• PixelCraft: A Multi-Agent System for High-Fidelity Visual Reasoning on Structured Images
• UFO: A UI-Focused Agent for Windows OS Interaction
• Multi-Agents Based on Large Language Models for Knowledge-based Visual Question Answering
• Embodied Multi-Modal Agent trained by an LLM from a Parallel TextWorld
• Two Heads Are Better Than One: Collaborative LLM Embodied Agents for Human-Robot Interaction
• LongVideoAgent: Multi-Agent Reasoning with Long Videos

Other Domains

• Human-level play in Diplomacy by combining language models with strategic reasoning
• CulturePark: Boosting Cross-cultural Understanding in Large Language Models
• Beyond Human Translation: Multi-Agent Collaboration for Translating Ultra-Long Literary Texts
• FanCric: Multi-Agentic Framework for Crafting Fantasy 11 Cricket Teams
• Can Large Language Models Grasp Legal Theories? Enhance Legal Reasoning with Multi-Agent Collaboration
• SQL-of-Thought: Multi-agentic Text-to-SQL with Guided Error Correction
• CoMet: Metaphor-Driven Covert Communication for Multi-Agent Language Games
• Richelieu: Self-Evolving LLM-Based Agents for AI Diplomacy
• Large Language Model for Participatory Urban Planning
• Large language model empowered participatory urban planning
• Agents on the Bench: Large Language Model Based Multi Agent Framework for Trustworthy Digital Justice
• AgentCourt: Simulating Court with Adversarial Evolvable Lawyer Agents
• MiniFed : Integrating LLM-based Agentic-Workflow for Simulating FOMC Meeting
• TRIZ Agents: A Multi-Agent LLM Approach for TRIZ-Based Innovation
• Multi-Agent Autonomous Driving Systems with Large Language Models: A Survey of Recent Advances
• KoMA: Knowledge-driven Multi-agent Framework for Autonomous Driving with Large Language Models
• A Multi-Agent Conversational Recommender System
• Enhancing Supermarket Robot Interaction: A Multi-Level LLM Conversational Interface for Handling Diverse Customer Intents
• Challenges Faced by Large Language Models in Solving Multi-Agent Flocking
• Scalable Multi-Robot Collaboration with Large Language Models: Centralized or Decentralized Systems?
• MALMM: Multi-Agent Large Language Models for Zero-Shot Robotics Manipulation
• ControlAgent: Automating Control System Design via Novel Integration of LLM Agents and Domain Expertise
• AgenticPay: A Multi-Agent LLM Negotiation System for Buyer-Seller Transactions

Evaluation & Model Improvement

• Wisdom of the Silicon Crowd: LLM Ensemble Prediction Capabilities Match Human Crowd Accuracy
• Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference
• Are More LLM Calls All You Need? Towards Scaling Laws of Compound Inference Systems
• Evolutionary Optimization of Model Merging Recipes
• Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models
• Constitutional AI: Harmlessness from AI Feedback
• On scalable oversight with weak LLMs judging strong LLMs
• ChatEval: Towards Better LLM-based Evaluators through Multi-Agent Debate
• RouteLLM: An Open-Source Framework for Cost-Effective LLM Routing
• Agent-as-a-Judge: Evaluate Agents with Agents
• Adversarial Multi-Agent Evaluation of Large Language Models through Iterative Debates
• MALT: Improving Reasoning with Multi-Agent LLM Training
• Why Do Multi-Agent LLM Systems Fail?
• Improving LLM Reasoning with Multi-Agent Tree-of-Thought Validator Agent
• Your Agents Are Aging Too: Agent Lifespan Engineering for Deployed Systems
• Share More, Search Less: Collaborative Parallel Thinking for Efficient Test-Time Scaling
• ResearchMath-14K: Scaling Research-Level Mathematics via Agents
• The Decrypto Benchmark for Multi-Agent Reasoning and Theory of Mind
• Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL
• Probe by Gaming: A Game-based Benchmark for Assessing Conceptual Knowledge in LLMs
• The Landscape of Agentic Reinforcement Learning for LLMs: A Survey
• Stance Detection with Collaborative Role-Infused LLM-Based Agents
• Content Knowledge Identification with Multi-Agent Large Language Models (LLMs)
• Why Solving Multi-agent Path Finding with Large Language Model has not Succeeded Yet
• Reliable Decision-Making for Multi-Agent LLM Systems
• Collab-Overcooked: Benchmarking and Evaluating Large Language Models as Collaborative Agents
• AutoDefense: Multi-Agent LLM Defense against Jailbreak Attacks
• AlignUSER: Human-Aligned LLM Agents via World Models for Recommender System Evaluation
• Beyond Offline A/B Testing: Context-Aware Agent Simulation for Recommender System Evaluation
• PerceptUI: LLM Agents as Human-Aligned Synthetic Users for UI/UX Evaluation

Social Simulation & Agent Societies

• TwinMarket: A Scalable Behavioral and Social Simulation for Financial Markets
• Generative Agents: Interactive Simulacra of Human Behavior
• SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents
• Scaling Instructable Agents Across Many Simulated Worlds
• Scaling Synthetic Data Creation with 1,000,000,000 Personas
• Cooperate or Collapse: Emergence of Sustainable Cooperation in a Society of LLM Agents
• From Text to Life: On the Reciprocal Relationship between Artificial Life and Large Language Models
• Mindstorms in Natural Language-Based Societies of Mind
• Agents' Room: Narrative Generation through Multi-step Collaboration
• GenSim: A General Social Simulation Platform with Large Language Model based Agents
• Large Language Models can Achieve Social Balance
• Cultural Evolution of Cooperation among LLM Agents
• SDPO: Segment-Level Direct Preference Optimization for Social Agents
• AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents
• OASIS: Open Agent Social Interaction Simulations with One Million Agents
• LLM-Based Social Simulations Require a Boundary
• Social World Models
• Simulating Human-like Daily Activities with Desire-driven Autonomy
• Implicit Behavioral Alignment of Language Agents in High-Stakes Crowd Simulations
• Greater than the Sum of its Parts: Exploring Social Influence of Multi-Agents
• Multi-Agents Are Social Groups: Investigating Social Influence of Multiple Agents in Human–Agent Interactions
• Digital Red Queen: Adversarial Program Evolution in Core War with LLMs
• Distributional AGI Safety: Governance Trade-offs in Multi-Agent Systems Under Adversarial Pressure
• Agentopia: Long-Term Life Simulation and Learning in Agent Societies
• CitySim: Modeling Urban Behaviors and City Dynamics with Large-Scale LLM-Driven Agent Simulation
• From Economic Agents to Agentic Economies: A Systems Blueprint for Economic World Models
• A game theory for foundation models shows new paths to rational cooperation through similarity inference Link
• MatrAIx: Simulating the World with 8.3 Billion Persona Agents LINK
• Business Arena: Benchmarking LLM Agents in a Realistic Marketplace LINK
• Poor Man's Agentic Modeling: Simulating Large LLM-Agent Societies on a Laptop LINK

Workflow, Architecture & Agent Design

• AllHands: Ask Me Anything on Large-scale Verbatim Feedback via Large Language Models
• AgentInstruct: Toward Generative Teaching with Agentic Flows
• SciAgents: Automating scientific discovery through multi-agent intelligent graph reasoning
• Minstrel: Structural Prompt Generation with Multi-Agents Coordination for Non-AI Experts
• AFlow: Automating Agentic Workflow Generation
• Agents Thinking Fast and Slow: A Talker-Reasoner Architecture
• DynaSaur: Large Language Agents Beyond Predefined Actions
• LLMs as Method Actors: A Model for Prompt Engineering and Architecture
• Proposer-Agent-Evaluator(PAE): Autonomous Skill Discovery For Foundation Model Internet Agents
• Automated Design of Agentic Systems
• The Fellowship of the LLMs: Multi-Agent Workflows for Synthetic Preference Optimization
• Multi-agent Architecture Search via Agentic Supernet
• Talk Structurally, Act Hierarchically: A Collaborative Framework for LLM Multi-Agent Systems
• When One LLM Drools, Multi-LLM Collaboration Rules
• Enhancing Reasoning with Collaboration and Memory
• Diffusion Augmented Agents: A Framework for Efficient Exploration and Transfer Learning
• Model Swarms: Collaborative Search to Adapt LLM Experts via Swarm Intelligence
• MDocAgent: A Multi-Modal Multi-Agent Framework for Document Understanding
• Recon-Act: A Self-Evolving Multi-Agent Browser-Use System via Web Reconnaissance, Tool Generation, and Task Execution
• Build Your Personalized Research Group: A Multiagent Framework for Continual and Interactive Science Automation
• Human-Agent Collaborative Paper-to-Page Crafting for Under $0.1
• Search Self-play: Pushing the Frontier of Agent Capability without Supervision
• Communication to Completion: Modeling Collaborative Workflows with Intelligent Multi-Agent Communication
• ComProScanner: A multi-agent based framework for composition-property structured data extraction from scientific literature
• LLM Voting: Human Choices and AI Collective Decision Making
• Planning with Multi-Constraints via Collaborative Language Agents
• An Electoral Approach to Diversify LLM-based Multi-Agent Collective Decision-Making
• Interactive Speculative Planning: Enhance Agent Efficiency through Co-design of System and User Interface
• Prompt Engineering Through the Lens of Optimal Control
• A Cooperative Multi-Agent Framework for Zero-Shot Named Entity Recognition
• Many Heads Are Better Than One: Improved Scientific Idea Generation by A LLM-Based Multi-Agent System
• PlotGen: Multi-Agent LLM-based Scientific Data Visualization via Multimodal Feedback
• Enhancing Anomaly Detection in Financial Markets with an LLM-based Multi-Agent Framework

Science

• GenoTEX: An LLM Agent Benchmark for Automated Gene Expression Data Analysis
• GenoMAS: A Multi-Agent Framework for Scientific Discovery via Code-Driven Gene Expression Analysis
• aiXiv: A Next-Generation Open Access Ecosystem for Scientific Discovery Generated by AI Scientists
• ScientistOne: Towards Human-Level Autonomous Research via Chain-of-Evidence
• AutoScientists: Self-Organizing Agent Teams for Long-Running Scientific Experimentation
• The AI Scientist: The world's first AI system for automating scientific research
• Deep Research in Physical Sciences: A Multi-Agent Framework and Comprehensive Benchmark

Alignment

• The Alignment Waltz: Jointly Training Agents to Collaborate for Safety
• MACPO: Weak-to-Strong Alignment via Multi-Agent Contrastive Preference Optimization

Reinforcement Learning

• MAPoRL: Multi-Agent Post-Co-Training for Collaborative Large Language Models with Reinforcement Learning
• YOLO-MARL: You Only LLM Once for Multi-agent Reinforcement Learning
• LLM-based Multi-Agent Reinforcement Learning: Current and Future Directions
• Collaborative Multi-Agent Test-Time Reinforcement Learning for Reasoning
• Scaling Multiagent Systems with Process Rewards

Surveys

• Large Language Model based Multi-Agents: A Survey of Progress and Challenges
• A survey on LLM-based multi-agent systems: workflow, infrastructure, and challenges
• Multi-Agent Collaboration Mechanisms: A Survey of LLMs
• A Survey on LLM-based Multi-Agent System: Recent Advances and New Frontiers in Application
• A Survey on Large Language Model based Autonomous Agents
• Large Language Models: A Survey
• LLM Multi-Agent Systems: Challenges and Open Problems
• Multi-Agent Coordination across Diverse Applications: A Survey
• LLM-based Multi-Agent Systems: Techniques and Business Perspectives
• Small LLMs Are Weak Tool Learners: A Multi-LLM Agent

Others

• PosterGen: Aesthetic-Aware Paper-to-Poster Generation via Multi-Agent LLMs
• MetaGPT: Meta Programming for Multi-Agent Collaborative Framework
• Multi-Agent Deep Research: Training Multi-Agent Systems with M-GRPO

Finance

• TwinMarket: A Scalable Behavioral and Social Simulation for Financial Markets
• HedgeAgents: A Balanced-aware Multi-agent Financial Trading System

Agents for Research

• Towards Collaborative Autonomous Research

Training

• Latent Collaboration in Multi-Agent Systems
• Recursive Multi-Agent Systems

---

Citations

In the arxiv_bibtex.bib file, you can find the bibtex citations for all the papers in this repository.

Cite This List

If you find this resource useful in your research, please cite it as follows:

BibTeX:

@misc{gomez2024awesome,
  author       = {Gomez, Kye},
  title        = {Awesome Multi-Agent Papers: A Compilation of the Best Multi-Agent Research},
  year         = {2024},
  publisher    = {GitHub},
  journal      = {GitHub Repository},
  howpublished = {\url{https://github.com/kyegomez/awesome-multi-agent-papers}},
  note         = {Maintained by the Swarms Team. Contact: kye@swarms.world}
}

APA:

Gomez, K. (2024). Awesome Multi-Agent Papers: A Compilation of the Best Multi-Agent Research. GitHub. https://github.com/kyegomez/awesome-multi-agent-papers

MLA:

Gomez, Kye. "Awesome Multi-Agent Papers: A Compilation of the Best Multi-Agent Research." GitHub, 2024, https://github.com/kyegomez/awesome-multi-agent-papers.

For questions or collaborations, contact: kye@swarms.world

Contributing

Have a multi-agent paper that isn’t on the list? We welcome your contributions! Please open a Pull Request (PR) to add new papers and help us maintain this comprehensive and up-to-date resource for the multi-agent research community. By contributing, you enable others—especially newcomers—to access the latest research in a single, centralized repository. Thank you for helping the community grow!

License

This project is licensed under the Apache License 2.0.