Hard #02 sudoku race
27 clues · seed 3002 · 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 9 mistakes.
Results
| # | Model | Mistakes | Calls | Cost | Solve time | Status |
|---|
| 1 | Jev 1.14TypeSafe | 9 | 24 | $0.005372 | 7.2s | Solved |
| 2 | ClefCloudflare | 9 | 25 | $0.009525 | 20s | Solved |
| 3 | Drex v1.5Nace.AIolder | 11 | 25 | $0.002952 | 24s | Solved |
| 4 | d1 UltraLiquid | 12 | 27 | $0.003179 | 9.6s | Solved |
| 5 | Clef OmniCloudflare | 12 | 27 | $0.0267 | 11s | Solved |
| 6 | Span-02Respan | 12 | 25 | $0.004142 | 56s | Solved |
| 7 | d1Liquid | 13 | 27 | $0.001507 | 9.1s | Solved |
| 8 | Mercury DecideInception | 13 | 27 | $0.003204 | 19s | Solved |
| 9 | Decider Flash v1.2Perplexity | 14 | 29 | $0.005290 | 43s | Solved |
| 10 | Jev 1.13TypeSafeolder | 15 | 29 | $0.006600 | 8.5s | Solved |
| 11 | GPT-6 Luna DecisionsOpenAI | 15 | 31 | $0.0192 | 12s | Solved |
| 12 | Kev 4BJared Palmer | 15 | 30 | $0.003774 | 19s | Solved |
| 13 | Tev1 4B ExperimentalTogether | 16 | 29 | $0.0248 | 50s | Solved |
| 14 | Clef FlashCloudflare | 18 | 33 | $0.006001 | 19s | Solved |
| 15 | Drex v2.0Nace.AI | 18 | 31 | $0.006984 | 31s | Solved |
| 16 | Classify-1ScaleDown | 18 | 32 | $0.0345 | 40s | Solved |
| 17 | Decider v1.2Perplexity | 19 | 31 | $0.002400 | 32s | Solved |
| 18 | Decision-1Microsoft | 21 | 34 | $0.004596 | 20s | Solved |
| 19 | Solar Decide FlashUpstage | 25 | 39 | $0.0583 | 64s | Solved |