SUDOKU BENCHMARK ยท STEADY

gpt-oss 20B on Steady grid 1

Solved: 53 of 53 blank cells right. It said it was 100% sure. All of gpt-oss 20B's answers.

RESULTSolved
CELLS RIGHT53 / 53
SAID100% sure
MODEL TIME55.8 min
COSTlocal, no list price
TOKENS IN / OUT333 / 16,661

5 given 5 filled correctly 53 wrong, with the right digit 3 left empty

The end of its reply

<grid>
243158697
815976324
976234518
621487935
459321876
387569241
598743162
164892753
732615489
</grid>
<confidence>100</confidence>

Grid steady-0 of set v1, 28 givens. Puzzle as one line: ..3..8.97.........9...3..18..14..93.45.....76.87..92..59..4...2.........73.6..4..