Hard #01 sudoku race
27 clues · seed 3001 · 19 models · suite v1
19 AI decision models raced to solve this hard sudoku (27 clues), one digit at a time. Span-02 by Respan came first with 7 mistakes.
Results
| # | Model | Mistakes | Calls | Cost | Solve time | Status |
|---|
| 1 | Span-02Respan | 7 | 23 | $0.003615 | 51s | Solved |
| 2 | Decider Flash v1.2Perplexity | 9 | 24 | $0.004222 | 35s | Solved |
| 3 | Jev 1.14TypeSafe | 10 | 24 | $0.005167 | 7.5s | Solved |
| 4 | Jev 1.13TypeSafeolder | 11 | 26 | $0.006102 | 6.7s | Solved |
| 5 | d1 UltraLiquid | 13 | 29 | $0.003553 | 12s | Solved |
| 6 | Drex v2.0Nace.AI | 14 | 29 | $0.006132 | 28s | Solved |
| 7 | Classify-1ScaleDown | 14 | 28 | $0.0299 | 35s | Solved |
| 8 | Clef FlashCloudflare | 15 | 31 | $0.006458 | 29s | Solved |
| 9 | Decision-1Microsoft | 16 | 32 | $0.004370 | 19s | Solved |
| 10 | d1Liquid | 17 | 30 | $0.001774 | 15s | Solved |
| 11 | ClefCloudflare | 17 | 31 | $0.009269 | 24s | Solved |
| 12 | Mercury DecideInception | 18 | 32 | $0.004126 | 16s | Solved |
| 13 | Kev 4BJared Palmer | 19 | 32 | $0.004187 | 22s | Solved |
| 14 | Drex v1.5Nace.AIolder | 20 | 34 | $0.004350 | 36s | Solved |
| 15 | Tev1 4B ExperimentalTogether | 20 | 35 | $0.0333 | 62s | Solved |
| 16 | Clef OmniCloudflare | 21 | 34 | $0.0355 | 14s | Solved |
| 17 | Decider v1.2Perplexity | 23 | 37 | $0.002989 | 48s | Solved |
| 18 | Solar Decide FlashUpstage | 23 | 37 | $0.0555 | 59s | Solved |
| 19 | GPT-6 Luna DecisionsOpenAI | 19 | 23 | $0.0176 | 9.2s | Error |