Jev 1.13
olderTypeSafe · typesafe/jev-1.13 · added 20 Oct 2026
Newest version: Jev 1.14. Ranks on this page count every version, older ones included.
Jev 1.13 by TypeSafe, an older version of Jev 1.14, ranks #5 of 19 AI decision models overall on DecisionBench: #4 of 19 on Sudoku, #10 of 19 on Rulebook, #8 of 19 on Consistency, #8 of 19 on Odds. View on OpenRouter.
- Overall
- #5 of 19
- Overall score
- 68 / 100
- Sudoku
- #4 of 19
- Rulebook
- #10 of 19
- Consistency
- #8 of 19
- Odds
- #8 of 19
Logical reasoningSudoku#4
- Accuracy
- 82.4%
- Solved
- 20/20
- Mistakes / puzzle
- 10.8
- Accuracy
- 92.8%
- Confidence gap
- 0.030
- Cost / 1k tickets
- $0.0409
- Score
- 99.3%
- Same if re-asked
- 96.7%
- Obeyed planted text
- 0.0%
- Score
- 39.1%
- Value captured
- 61.5%
- Odds within 2×
- 16.7%
On each benchmark the best model scores 100 and the worst 0. The overall score averages all 4, so every benchmark counts the same. How ranking works.