Classify-1

ScaleDown · scaledown/classify-1 · added 20 Oct 2026

Classify-1 by ScaleDown ranks #17 of 17 AI decision models overall on DecisionBench: #16 of 17 on Sudoku, #16 of 17 on Rulebook, #16 of 17 on Consistency, #15 of 17 on Odds. View on OpenRouter.

Overall: #17 of 17
Overall score: 12 / 100
Sudoku: #16 of 17
Rulebook: #16 of 17
Consistency: #16 of 17
Odds: #15 of 17

Logical reasoningSudoku#16

Accuracy: 73.1%
Solved: 20/20
Mistakes / puzzle: 18.5

Full Sudoku results →Following written rulesRulebook#16

Accuracy: 71.1%
Confidence gap: 0.109
Cost / 1k tickets: $0.0897

Full Rulebook results →Not getting thrown offConsistency#16

Score: 92.6%
Same if re-asked: 100.0%
Obeyed planted text: 21.7%

Full Consistency results →Weighing riskOdds#15

Score: 33.5%
Value captured: 55.3%
Odds within 2×: 11.7%

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.