MultiAgentBench : Evaluating the Collaboration and Competition of LLM agents
The DOI matches its Crossref record; 4 of the 4 references it cites were found in a registry or an index.
Abstract
Large Language Models (LLMs) have shown remarkable capabilities as autonomous agents, yet existing benchmarks either focus on single-agent tasks or are confined to narrow domains, failing to capture the dynamics of multi-agent coordination and competition. In this paper, we introduce MultiAgentBench, a comprehensive benchmark designed to evaluate LLM-based multi-agent systems across diverse, interactive scenarios. Our framework measures not only task completion but also the quality of collaboration and competition using novel, milestone-based key performance indicators. Moreover, we evaluate various coordination protocols (including star, chain, tree, and graph topologies) and innovative strategies such as group discussion and cognitive planning. Notably, gpt-4o-mini reaches the average highest task score, graph structure performs the best among coordination protocols in the research scenario, and cognitive planning improves milestone achievement rates by 3%. Code and datasets are public available at https://github.com/MultiagentBench/MARBLE.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Generative Agents: Interactive Simulacra of Human Behavior
Park, J. S., O'Brien, J. C., Cai, C. J., Morris, M. R., Liang, P., & Bernstein, M. S. (2023) · arXiv · doi:10.48550/arxiv.2304.03442 · Cited by 2,173
MDAgents: an adaptive collaboration of LLMs for medical decision-making
Kim, Y., Park, C., Jeong, H., Chan, Y., Xu, X., McDuff, D., Lee, H., Ghassemi, M., Breazeal, C., & Park, H. (2024) · Advances in Neural Information Processing Systems 37 · doi:10.52202/079017-2522 · Cited by 81
Ishibashi, Y., & Nishimura, Y. (2024) · arXiv · doi:10.48550/arxiv.2404.02183 · Cited by 8
Ke, Y. H., Yang, R., Lie, S. A., Lim, T. X. Y., Abdullah, H. R., Ting, D. S. W., & Liu, N. (2024) · arXiv · doi:10.48550/arxiv.2401.14589 · Cited by 5
Cited by
Open Challenges in Multi-Agent Security: Towards Secure Systems of Interacting AI Agents
de Witt, C. S., Krawiecka, K., Krawczuk, I., Hagag, B., Anderson, W. L., Belcak, P., Bucknall, B., Cai, X., Chopra, A., Cohen, D., et al. (2025) · arXiv · doi:10.48550/arxiv.2505.02077 · Cited by 108
Heterogeneous Swarms: Jointly Optimizing Model Roles and Weights for Multi-LLM Systems
Feng, S., Wang, Z., Goyal, P., Wang, Y., Shi, W., Xia, H., Palangi, H., Zettlemoyer, L., Tsvetkov, Y., Lee, C.-Y., et al. (2025) · arXiv · doi:10.48550/arxiv.2502.04510 · Cited by 18
Self-citationLightweight LLM Agent Memory with Small Language Models
Zhang, J., Zhang, C., Chen, S., Huang, Z., Zheng, P., Wang, Z., Guo, P., Mo, F., Bae, S.-H., Zou, J., et al. (2026) · Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2026.acl-long.588 · Cited by 35
Self-citationMemory in LLM-Based Multi-agent Systems: Mechanisms, Challenges, and Collective Intelligence
Wu, S., & Shu, K. (2026) · Lecture Notes in Computer Science · doi:10.1007/978-981-92-1468-6_10 · Cited by 2
SidConArena: An Environment Evaluating Agents in Open-Ended,Positive-Sum Bargaining Game
Feng, Y., Chen, Y., & He, T. (2026) · arXiv · doi:10.48550/arxiv.2606.27397 · Cited by 1
Metadata
- Authors
- Zhu, K., Du, H., Hong, Z., Yang, X., Guo, S., Wang, Z., Wang, Z., Qian, C., Tang, R., Ji, H., & You, J.
- Venue
- Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
- Published
- 2025
- Pages
- 8580-8622
- Cited by
- 47
- References
- 4
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.