SUDOKU BENCHMARK · LOCAL, NO THINKING

Ministral 3 8B

Mistral · open weights · released 2025-12-02 · route ministral-3:8b-instruct-2512-q4_K_M · Model profile · Chess results

Results

Score
095%: 0 to 20 · 7 of 42 grids so far
Grids solved
0 of 7
Blank cells filled correctly
12.4%
Wrong cells / empty cells / changed a given / no grid
0 / 0 / 7 / 0
Broke the rules (complete grid repeating a digit)
0
Near misses (three cells or fewer wrong)
0
Average stated confidence
100%
Average confidence when it had not solved it
100%
Sure but wrong (90% or more, unsolved)
7
Calibration (Brier, lower is better)
1.000
API-equivalent cost per grid
n/a
API-equivalent cost per solved grid
n/a
Seconds per grid
20
Tokens per grid (in / out)
304 / 108
Reasoning tokens per grid
n/a
Retries for an unreadable reply
0
Aborted on our side (not counted)
0

By tier

TierNeedsSolvedOutput tokensSeconds
CozySingles only0 of 110820
GentleScanning0 of 110822
SteadyLocked candidates0 of 110823
TrickyHidden pairs and triples0 of 110820
DeviousWings and fish0 of 110824
ExpertDeep wings and fish0 of 110811
BeyondNo human technique finishes it0 of 110816

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 cell629.7%
Scanning13713.9%
Locked candidates468.7%
Hidden pairs and triples5311.3%
Wings and fish229.1%
Deep wings and fish3420.6%
Search3312.1%

Every answer