MINES BENCHMARK · REASONING MEDIUM

GPT-6 Sol

Openai · released 2026-09-22 · route gpt-6-sol · Model profile · Chess results · Sudoku results · FreeCell results · Hustle Sim results · Elemental Armory results · Micro Mouse results

Results

Score
5095%: 23 to 77 · 4 of 24 base boards so far
Boards won
2 of 4
Safe cells cleared beyond the opening
68.2%
Won / hit a mine / unfinished / unreadable
2 / 2 / 0 / 0
Ended for changing nothing five turns running
0
Reached the turn cap
0
Flags left on safe cells
2 in 1 boards
Illegal actions
0 of 470 (0.0%)
Wasted actions (already open or flagged)
37 of 470 (7.9%)
Unreadable replies
0
Turns per board
7.8
Requests per board
7.8
API-equivalent cost per win
$0.636
Tokens per board (in / out)
153,227 / 24,046
Reasoning tokens per board
20,754
API-equivalent cost per board
$0.318
Seconds per board
551.1
Replies cut off at the length limit
0.0%
Aborted on our side (not counted)
0

By tier

TierBoard, minesWonClearedTurnsOutput tokensSeconds
Beginner9×9, 101 of 1100%5.05,767127.9
Intermediate16×16, 401 of 1100%6.023,704472.8
Expert30×16, 990 of 152%6.023,114764.5
Master30×24, 2000 of 121%14.043,599839.1

Cells by the technique that proves them

Each safe cell covered at the start is tagged with the simplest deduction that first proves it safe, given everything a perfect player could open before it.

TechniqueCellsOpened
Cascade from an empty cell45858.1%
One number56940.8%
Two numbers together6839.7%
The mine count40.0%

Every board