Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Multi-Agent Reasoning research benchmarks · SOTA2 Research

Multi-Agent Reasoning

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

85.78Accuracy

9

May 20, 2026

91.02Accuracy

9

May 20, 2026

73.66Accuracy

9

May 20, 2026

198Calls

9

Feb 26, 2026

30Calls

9

Feb 26, 2026

Reasoning Benchmarks Cooperative AutoGen framework (test)

MARSHAL (Generalist, 8B)

83.58Overall Accuracy

2

Feb 26, 2026

Reasoning Benchmarks Competitive MAD framework (test)

MARSHAL (Generalist, 8B)

0.851Average Score

2

Feb 26, 2026

Popular datasets

Reasoning Benchmarks MMLU AIME GPQA UltraFeedback

Follow for updates

Join SOTA2 Contact