SUDOKU BENCHMARK · REASONING MEDIUM

GPT-6.1 Terra

Openai · released 2026-10-26 · route gpt-6.1-terra · Model profile · Chess results

Results

Score
795%: 0 to 17
Grids solved
3 of 42
Blank cells filled correctly
10.2%
Wrong cells / empty cells / changed a given / no grid
2 / 24 / 1 / 12
Broke the rules (complete grid repeating a digit)
2
Near misses (three cells or fewer wrong)
0
Average stated confidence
12%
Average confidence when it had not solved it
5%
Sure but wrong (90% or more, unsolved)
2
Calibration (Brier, lower is better)
0.048
API-equivalent cost per grid
$0.015
API-equivalent cost per solved grid
$0.208
Seconds per grid
26
Tokens per grid (in / out)
39,516 / 1,849
Reasoning tokens per grid
1,330
Retries for an unreadable reply
19
Aborted on our side (not counted)
0

By tier

TierNeedsSolvedOutput tokensSeconds
CozySingles only3 of 63,64941
GentleScanning0 of 61,94825
SteadyLocked candidates0 of 61,91224
TrickyHidden pairs and triples0 of 61,18230
DeviousWings and fish0 of 62,00424
ExpertDeep wings and fish0 of 61,44721
BeyondNo human technique finishes it0 of 679914

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 cell29762.6%
Scanning4836.6%
Locked candidates2290.4%
Hidden pairs and triples740.0%
Wings and fish1435.6%
Deep wings and fish1330.0%
Search2160.0%

Every answer

Cozy #1No readable grid 0 of 50 cells0% sure · 4,557 tokens · $0.041Cozy #2Left cells empty 5 of 48 cells0% sure · 1,373 tokens · $0.013Cozy #3Solved 49 of 49 cells100% sure · 6,589 tokens · $0.041Cozy #4No readable grid 0 of 50 cells0% sure · 1,251 tokens · $0.012Cozy #5Solved 49 of 49 cells100% sure · 2,738 tokens · $0.016Cozy #6Solved 39 of 39 cells100% sure · 5,384 tokens · $0.032Gentle #1No readable grid 0 of 56 cells0% sure · 981 tokens · $0.013Gentle #2Left cells empty 0 of 51 cells0% sure · 801 tokens · $0.0073Gentle #3Left cells empty 13 of 52 cells0% sure · 2,691 tokens · $0.017Gentle #4Left cells empty 0 of 51 cells0% sure · 2,114 tokens · $0.014Gentle #5Left cells empty 0 of 56 cells0% sure · 454 tokens · $0.0047Gentle #6Left cells empty 11 of 52 cells0% sure · 4,646 tokens · $0.030Steady #1Left cells empty 0 of 53 cells0% sure · 569 tokens · $0.015Steady #2Left cells empty 0 of 53 cells0% sure · 489 tokens · $0.0049Steady #3Changed a given digit 3 of 49 cells0% sure · 3,972 tokens · $0.023Steady #4No readable grid 0 of 54 cells0% sure · 1,664 tokens · $0.014Steady #5Left cells empty 6 of 53 cells0% sure · 4,073 tokens · $0.026Steady #6Left cells empty 0 of 54 cells0% sure · 707 tokens · $0.0067Tricky #1No readable grid 0 of 55 cells0% sure · 1,380 tokens · $0.018Tricky #2No readable grid 0 of 57 cells0% sure · 1,287 tokens · $0.022Tricky #3Left cells empty 0 of 57 cells0% sure · 1,308 tokens · $0.012Tricky #4No readable grid 0 of 54 cells0% sure · 1,650 tokens · $0.014Tricky #5Left cells empty 0 of 55 cells0% sure · 570 tokens · $0.0071Tricky #6No readable grid 0 of 56 cells0% sure · 896 tokens · $0.010Devious #1No readable grid 0 of 55 cells0% sure · 2,097 tokens · $0.019Devious #2Left cells empty 0 of 55 cells0% sure · 760 tokens · $0.0062Devious #3Wrong cells 22 of 49 cells100% sure · 4,497 tokens · $0.026Devious #4Wrong cells 24 of 40 cells100% sure · 2,245 tokens · $0.014Devious #5Left cells empty 6 of 47 cells0% sure · 1,929 tokens · $0.012Devious #6Left cells empty 0 of 54 cells0% sure · 498 tokens · $0.0057Expert #1Left cells empty 0 of 58 cells0% sure · 525 tokens · $0.0058Expert #2No readable grid 0 of 53 cells0% sure · 5,255 tokens · $0.036Expert #3Left cells empty 0 of 54 cells0% sure · 845 tokens · $0.0067Expert #4Left cells empty 0 of 57 cells0% sure · 486 tokens · $0.0056Expert #5No readable grid 0 of 58 cells0% sure · 968 tokens · $0.011Expert #6Left cells empty 0 of 54 cells0% sure · 604 tokens · $0.0055Beyond #1No readable grid 0 of 60 cells0% sure · 894 tokens · $0.011Beyond #2Left cells empty 0 of 57 cells0% sure · 485 tokens · $0.0049Beyond #3Left cells empty 0 of 56 cells0% sure · 1,291 tokens · $0.014Beyond #4Left cells empty 0 of 57 cells0% sure · 1,065 tokens · $0.012Beyond #5Left cells empty 0 of 59 cells0% sure · 523 tokens · $0.0087Beyond #6Left cells empty 0 of 57 cells0% sure · 538 tokens · $0.0059