Medium #01 sudoku race
32 clues · seed 2001 · 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 5 mistakes.
Results
| # | Model | Mistakes | Calls | Cost | Solve time | Status |
|---|
| 1 | Jev 1.13TypeSafeolder | 5 | 20 | $0.004057 | 8.5s | Solved |
| 2 | Clef FlashCloudflare | 5 | 20 | $0.002929 | 11s | Solved |
| 3 | d1 UltraLiquid | 6 | 20 | $0.002170 | 6.7s | Solved |
| 4 | Clef OmniCloudflare | 7 | 22 | $0.0196 | 7.7s | Solved |
| 5 | d1Liquid | 7 | 22 | $0.001147 | 9.1s | Solved |
| 6 | Drex v1.5Nace.AIolder | 8 | 23 | $0.002465 | 15s | Solved |
| 7 | Drex v2.0Nace.AI | 8 | 21 | $0.003867 | 17s | Solved |
| 8 | Decision-1Microsoft | 9 | 23 | $0.002719 | 13s | Solved |
| 9 | Span-02Respan | 9 | 22 | $0.003447 | 49s | Solved |
| 10 | Jev 1.14TypeSafe | 10 | 23 | $0.005179 | 6.9s | Solved |
| 11 | Classify-1ScaleDown | 13 | 27 | $0.0276 | 22s | Solved |
| 12 | ClefCloudflare | 15 | 28 | $0.008855 | 21s | Solved |
| 13 | Tev1 4B ExperimentalTogether | 15 | 29 | $0.0258 | 31s | Solved |
| 14 | Kev 4BJared Palmer | 16 | 28 | $0.003410 | 18s | Solved |
| 15 | Decider v1.2Perplexity | 16 | 28 | $0.002042 | 27s | Solved |
| 16 | Decider Flash v1.2Perplexity | 16 | 25 | $0.004489 | 35s | Solved |
| 17 | Solar Decide FlashUpstage | 16 | 27 | $0.0374 | 44s | Solved |
| 18 | GPT-6 Luna DecisionsOpenAI | 21 | 31 | $0.0193 | 13s | Solved |
| 19 | Mercury DecideInception | 22 | 34 | $0.004255 | 16s | Solved |