MICRO MOUSE BENCHMARK · LOCAL, NO THINKING

Gemma 3 4B

Google · open weights · released 2025-03-12 · route gemma3:4b · Model profile · Chess results · Sudoku results · FreeCell results · Hustle Sim results · Elemental Armory results

Results

Score
095%: 0 to 28 · 1 of 14 base runs so far
Campaign score against the bot
n/a
Daily mazes faster than the median person
0 of 1
Reached the dock
0 of 1 mazes
Faster than the bot
0 of 1
Reached / out of time / out of replies
0 / 0 / 1
Share of the route covered, on average
10%
Wall hits per run
64.0
Drift boosts / pad boosts per run
0.0 / 0.0
Replies per run
71.0
Control segments per reply
2.03
Values moved into range
0
Segments cut at the one-second limit
139
Lines that were not segments
0
Replies with no readable segment
19.7%
Retries for an unreadable reply
14
Model minutes per run
335.3
Tokens per reply (in / out)
2,028 / 654
Reasoning tokens per reply
n/a
API-equivalent cost per run
n/a
Seconds per reply
283.3
Replies cut off at the length limit
1.2%
Aborted on our side (not counted)
2

By tier

TierAgainstRunsScoreReachedPaceWall hits / runTest pilot
Daily mazesmedian person100 of 1n/a64.048
Workbenchbot0n/a0 of 0n/an/a52
Cold Storagebot0n/a0 of 0n/an/a49
Fast Lanebot0n/a0 of 0n/an/a50
Night Shiftbot0n/a0 of 0n/an/a50

Every run