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

Mathematical Problem Solving research benchmarks – Page 9 · SOTA2 Research

Mathematical Problem Solving

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

MATH Counting & Probability (test)

474N

1

Jun 2, 2026

MATH Number Theory (test)

540N

1

Jun 2, 2026

Mismatched-wrong

71.98Pass@1

1

May 19, 2026

MathArena FA Total

60Correctness Ratio

1

May 11, 2026

MathArena FA HMMT 2025

11Correlation

1

May 11, 2026

MathArena FA CMIMC 2025

13Accuracy (Ratio)

1

May 11, 2026

MathArena FA BrUMO 2025

21Correlation Score

1

May 11, 2026

Winner (AIMO-3)

46Winner

1

Mar 31, 2026

Winner (AIMO-2)

34Winner Count

1

Mar 31, 2026

Winner (AIMO-1)

29Winner Score

1

Mar 31, 2026

T-Math 4 (test)

2.25Speedup

1

Feb 26, 2026

Erdős problems (All)

205Autonomous Resolution

1

Feb 26, 2026

DeepSeekMath-V2

98.3Points

1

Feb 26, 2026

DeepSeekMath-V2

73.8Points

1

Feb 26, 2026

—

—Primary metric

0

Apr 23, 2026

—

—Primary metric

0

Feb 26, 2026

—

—Primary metric

0

Feb 26, 2026

—

—Primary metric

0

Feb 26, 2026

Page 9 of 9

Popular datasets

AIME MATH AMC HMMT OlympiadBench

Follow for updates

Join SOTA2 Contact