Skip to content
decisionbench
OverallSudokuRulebookConsistency

Independent benchmark. Not affiliated with or endorsed by OpenRouter or any model provider. Company and model names are trademarks of their respective owners.

How ranking worksPrivacyTermsContact

© 2026 editid0

← Sudoku leaderboard

Jev 1.14

TypeSafe · typesafe/jev-1.14 · added 20 Oct 2026

Jev 1.14 by TypeSafe ranks #2 of 12 AI decision models on the DecisionBench Sudoku benchmark (logical reasoning), with accuracy 83.2%. View on OpenRouter.

Overall#2Sudoku#2Rulebook#6Consistency#7
Sudoku rank
#2 of 12
Accuracy
83.2%
Solved
20/20
Mistakes / puzzle
10.2
Solve time / puzzle
7.4s
Cost / puzzle
$0.005278

By difficulty

DifficultyRankAccuracySolvedMistakesSolve timeCost
Easy#192.8%5/53.24.8s$0.002945
Medium#484.5%5/59.06.8s$0.004795
Hard#184.1%5/510.27.4s$0.005438
Expert#575.9%5/518.211s$0.007936

Every race

PuzzleMistakesCallsSolve timeCostStatus
easy-012165.2s$0.002952Solved
easy-023164.7s$0.002901Solved
easy-036175.6s$0.003174Solved
easy-043164.6s$0.002916Solved
easy-052154.1s$0.002780Solved
medium-0110236.9s$0.005179Solved
medium-026216.1s$0.004360Solved
medium-038216.3s$0.004471Solved
medium-0410247.5s$0.005265Solved
medium-0511227.0s$0.004700Solved
hard-0110247.5s$0.005167Solved
hard-029247.2s$0.005372Solved
hard-0311257.5s$0.005610Solved
hard-0414278.3s$0.005845Solved
hard-057236.4s$0.005195Solved
expert-01223811s$0.009587Solved
expert-0213288.8s$0.006579Solved
expert-0315289.6s$0.006736Solved
expert-04203413s$0.008508Solved
expert-05213411s$0.008270Solved

Jev 1.14 across all benchmarks · How Sudoku works