SUDOKU BENCHMARK · REASONING MEDIUM

Claude Opus 5.5

Anthropic · released 2026-09-22 · route claude-opus-5-5 · Model profile · Chess results

Results

Score
10095%: 91 to 100
Grids solved
42 of 42
Blank cells filled correctly
100.0%
Wrong cells / empty cells / changed a given / no grid
0 / 0 / 0 / 0
Broke the rules (complete grid repeating a digit)
0
Near misses (three cells or fewer wrong)
0
Average stated confidence
99%
Average confidence when it had not solved it
n/a
Sure but wrong (90% or more, unsolved)
0
Calibration (Brier, lower is better)
0.000
API-equivalent cost per grid
$0.244
API-equivalent cost per solved grid
$0.244
Seconds per grid
101
Tokens per grid (in / out)
910 / 11,827
Reasoning tokens per grid
11,720
Retries for an unreadable reply
0
Aborted on our side (not counted)
0

By tier

TierNeedsSolvedOutput tokensSeconds
CozySingles only6 of 64,52141
GentleScanning6 of 68,29671
SteadyLocked candidates6 of 612,067102
TrickyHidden pairs and triples6 of 616,782143
DeviousWings and fish6 of 68,39874
ExpertDeep wings and fish6 of 611,761102
BeyondNo human technique finishes it6 of 620,962174

Cells by the technique they needed

Each blank cell is tagged with the simplest technique that fills it, given everything a solver could fill before it.

TechniqueCellsFilled correctly
Last free cell432100.0%
Scanning674100.0%
Locked candidates274100.0%
Hidden pairs and triples258100.0%
Wings and fish165100.0%
Deep wings and fish181100.0%
Search249100.0%

Every answer

Cozy #1Solved 50 of 50 cells99% sure · 6,155 tokens · $0.130Cozy #2Solved 48 of 48 cells98% sure · 4,320 tokens · $0.094Cozy #3Solved 49 of 49 cells99% sure · 4,891 tokens · $0.105Cozy #4Solved 50 of 50 cells98% sure · 4,224 tokens · $0.092Cozy #5Solved 49 of 49 cells98% sure · 4,713 tokens · $0.102Cozy #6Solved 39 of 39 cells97% sure · 2,821 tokens · $0.064Gentle #1Solved 56 of 56 cells99% sure · 11,483 tokens · $0.237Gentle #2Solved 51 of 51 cells99% sure · 7,892 tokens · $0.165Gentle #3Solved 52 of 52 cells99% sure · 7,432 tokens · $0.156Gentle #4Solved 51 of 51 cells99% sure · 6,695 tokens · $0.141Gentle #5Solved 56 of 56 cells99% sure · 9,477 tokens · $0.197Gentle #6Solved 52 of 52 cells99% sure · 6,795 tokens · $0.143Steady #1Solved 53 of 53 cells99% sure · 10,144 tokens · $0.210Steady #2Solved 53 of 53 cells99% sure · 8,562 tokens · $0.179Steady #3Solved 49 of 49 cells99% sure · 6,845 tokens · $0.144Steady #4Solved 54 of 54 cells99% sure · 13,500 tokens · $0.277Steady #5Solved 53 of 53 cells99% sure · 11,004 tokens · $0.227Steady #6Solved 54 of 54 cells99% sure · 22,349 tokens · $0.454Tricky #1Solved 55 of 55 cells99% sure · 9,881 tokens · $0.205Tricky #2Solved 57 of 57 cells99% sure · 38,164 tokens · $0.771Tricky #3Solved 57 of 57 cells99% sure · 19,667 tokens · $0.401Tricky #4Solved 54 of 54 cells98% sure · 9,107 tokens · $0.189Tricky #5Solved 55 of 55 cells99% sure · 7,272 tokens · $0.153Tricky #6Solved 56 of 56 cells99% sure · 16,603 tokens · $0.339Devious #1Solved 55 of 55 cells99% sure · 12,170 tokens · $0.251Devious #2Solved 55 of 55 cells99% sure · 11,121 tokens · $0.230Devious #3Solved 49 of 49 cells99% sure · 6,056 tokens · $0.128Devious #4Solved 40 of 40 cells98% sure · 4,464 tokens · $0.097Devious #5Solved 47 of 47 cells98% sure · 7,356 tokens · $0.154Devious #6Solved 54 of 54 cells99% sure · 9,224 tokens · $0.192Expert #1Solved 58 of 58 cells99% sure · 18,807 tokens · $0.383Expert #2Solved 53 of 53 cells99% sure · 6,811 tokens · $0.144Expert #3Solved 54 of 54 cells99% sure · 9,845 tokens · $0.204Expert #4Solved 57 of 57 cells98% sure · 14,467 tokens · $0.297Expert #5Solved 58 of 58 cells99% sure · 11,675 tokens · $0.241Expert #6Solved 54 of 54 cells98% sure · 8,959 tokens · $0.186Beyond #1Solved 60 of 60 cells99% sure · 15,536 tokens · $0.314Beyond #2Solved 57 of 57 cells99% sure · 30,356 tokens · $0.610Beyond #3Solved 56 of 56 cells99% sure · 26,961 tokens · $0.547Beyond #4Solved 57 of 57 cells99% sure · 14,079 tokens · $0.289Beyond #5Solved 59 of 59 cells99% sure · 13,090 tokens · $0.269Beyond #6Solved 57 of 57 cells99% sure · 25,750 tokens · $0.522