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

General Language Model Evaluation research benchmarks · SOTA2 Research

General Language Model Evaluation

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

Utility Set MMLU, BBH, TruthfulQA, TriviaQA, AlpacaEval

68.93MMLU

34

Apr 16, 2026

Aggregated 11-benchmark suite Math, Code, IF

74.9Average Accuracy

21

May 21, 2026

Comprehensive Evaluation Suite

50.7Overall Average Score

14

Jun 2, 2026

Average of 10 tasks

45.02Overall Performance

12

Feb 26, 2026

OlmoBaseEval HeldOut (LBPP, BBH, MMLU Pro, etc.)

33.7LBPP Score

10

Apr 7, 2026

Arena-Hard V2.0

7.03Win Rate

9

Feb 26, 2026

26.95WildBench Score

2

Feb 26, 2026

Popular datasets

Arena-Hard WildBench Average of 10 tasks OlmoBaseEval Utility Set

Follow for updates

Join SOTA2 Contact