Hard #05 sudoku race
27 clues · seed 3005 · 19 models · suite v1
19 AI decision models raced to solve this hard sudoku (27 clues), one digit at a time. Jev 1.14 by TypeSafe came first with 7 mistakes.
Results
| # | Model | Mistakes | Calls | Cost | Solve time | Status |
|---|
| 1 | Jev 1.14TypeSafe | 7 | 23 | $0.005195 | 6.4s | Solved |
| 2 | GPT-6 Luna DecisionsOpenAI | 12 | 27 | $0.0159 | 11s | Solved |
| 3 | Span-02Respan | 12 | 26 | $0.004358 | 58s | Solved |
| 4 | Jev 1.13TypeSafeolder | 14 | 29 | $0.006807 | 7.2s | Solved |
| 5 | d1 UltraLiquid | 14 | 26 | $0.003134 | 8.7s | Solved |
| 6 | Kev 4BJared Palmer | 14 | 29 | $0.003654 | 18s | Solved |
| 7 | Clef FlashCloudflare | 14 | 29 | $0.005413 | 19s | Solved |
| 8 | Mercury DecideInception | 15 | 28 | $0.003477 | 14s | Solved |
| 9 | Decision-1Microsoft | 16 | 29 | $0.003854 | 17s | Solved |
| 10 | Clef OmniCloudflare | 17 | 32 | $0.0326 | 13s | Solved |
| 11 | ClefCloudflare | 17 | 31 | $0.0115 | 23s | Solved |
| 12 | Solar Decide FlashUpstage | 17 | 32 | $0.0451 | 54s | Solved |
| 13 | Drex v1.5Nace.AIolder | 19 | 32 | $0.003825 | 29s | Solved |
| 14 | Drex v2.0Nace.AI | 19 | 34 | $0.007654 | 29s | Solved |
| 15 | d1Liquid | 21 | 33 | $0.002065 | 10s | Solved |
| 16 | Decider Flash v1.2Perplexity | 21 | 34 | $0.007441 | 55s | Solved |
| 17 | Tev1 4B ExperimentalTogether | 24 | 37 | $0.0339 | 40s | Solved |
| 18 | Decider v1.2Perplexity | 26 | 39 | $0.003231 | 84s | Solved |
| 19 | Classify-1ScaleDown | 29 | 42 | $0.0511 | 44s | Solved |