Medium #05 sudoku race
32 clues · seed 2005 · 19 models · suite v1
19 AI decision models raced to solve this medium sudoku (32 clues), one digit at a time. Jev 1.13 by TypeSafe came first with 3 mistakes.
Results
| # | Model | Mistakes | Calls | Cost | Solve time | Status |
|---|
| 1 | Jev 1.13TypeSafeolder | 3 | 19 | $0.003744 | 3.8s | Solved |
| 2 | ClefCloudflare | 4 | 20 | $0.008838 | 25s | Solved |
| 3 | Clef OmniCloudflare | 6 | 21 | $0.0198 | 8.9s | Solved |
| 4 | Span-02Respan | 6 | 20 | $0.003155 | 42s | Solved |
| 5 | Mercury DecideInception | 7 | 22 | $0.002565 | 11s | Solved |
| 6 | Decision-1Microsoft | 7 | 22 | $0.002599 | 13s | Solved |
| 7 | Drex v2.0Nace.AI | 7 | 22 | $0.004294 | 15s | Solved |
| 8 | d1 UltraLiquid | 9 | 25 | $0.002816 | 9.9s | Solved |
| 9 | Drex v1.5Nace.AIolder | 10 | 23 | $0.002519 | 27s | Solved |
| 10 | Jev 1.14TypeSafe | 11 | 22 | $0.004700 | 7.0s | Solved |
| 11 | GPT-6 Luna DecisionsOpenAI | 11 | 24 | $0.0137 | 12s | Solved |
| 12 | Clef FlashCloudflare | 11 | 27 | $0.005087 | 24s | Solved |
| 13 | Decider v1.2Perplexity | 11 | 25 | $0.001861 | 27s | Solved |
| 14 | Decider Flash v1.2Perplexity | 11 | 24 | $0.004481 | 35s | Solved |
| 15 | d1Liquid | 12 | 23 | $0.001201 | 8.0s | Solved |
| 16 | Solar Decide FlashUpstage | 16 | 30 | $0.0381 | 43s | Solved |
| 17 | Tev1 4B ExperimentalTogether | 18 | 30 | $0.0256 | 44s | Solved |
| 18 | Kev 4BJared Palmer | 20 | 32 | $0.004026 | 20s | Solved |
| 19 | Classify-1ScaleDown | 23 | 35 | $0.0391 | 33s | Solved |