gpt-oss 20B

Results

Score

2995%: 8 to 49 · 7 of 42 grids so far

Grids solved

2 of 7

Blank cells filled correctly

53.8%

Wrong cells / empty cells / changed a given / no grid

0 / 0 / 3 / 2

Broke the rules (complete grid repeating a digit)

0

Near misses (three cells or fewer wrong)

0

Average stated confidence

80%

Average confidence when it had not solved it

67%

Sure but wrong (90% or more, unsolved)

2

Calibration (Brier, lower is better)

0.400

API-equivalent cost per grid

n/a

API-equivalent cost per solved grid

n/a

Seconds per grid

3,050

Tokens per grid (in / out)

445 / 29,731

Reasoning tokens per grid

n/a

Retries for an unreadable reply

2

Aborted on our side (not counted)

110

By tier

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.