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-step mathematical reasoning research benchmarks · SOTA2 Research

Multi-step mathematical reasoning

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

72.8S1 Score

20

Feb 26, 2026

96.61Accuracy

14

May 28, 2026

MMLU-Pro (held-out test)

83.67Accuracy

14

May 28, 2026

90.84Accuracy

14

May 28, 2026

54Accuracy

5

Mar 17, 2026

Popular datasets

MMLU GSM8K DART We-Math

Follow for updates

Join SOTA2 Contact