Kev 4B

Jared Palmer · jaredpalmer/kev-4b · added 20 Oct 2026

Kev 4B by Jared Palmer ranks #14 of 17 AI decision models overall on DecisionBench: #12 of 17 on Sudoku, #10 of 17 on Rulebook, #15 of 17 on Consistency, #14 of 17 on Odds. View on OpenRouter.

Overall: #14 of 17
Overall score: 37 / 100
Sudoku: #12 of 17
Rulebook: #10 of 17
Consistency: #15 of 17
Odds: #14 of 17

Logical reasoningSudoku#12

Accuracy: 77.0%
Solved: 20/20
Mistakes / puzzle: 15.1

Full Sudoku results →Following written rulesRulebook#10

Accuracy: 89.4%
Confidence gap: 0.087
Cost / 1k tickets: $0.0255

Full Rulebook results →Not getting thrown offConsistency#15

Score: 93.1%
Same if re-asked: 100.0%
Obeyed planted text: 15.0%

Full Consistency results →Weighing riskOdds#14

Score: 34.7%
Value captured: 66.0%
Odds within 2×: 3.3%

Full Odds results →

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.