Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Maze Traversal on 10x10 Mazes (10 runs)

10Correct Count (out of 10)

gpt-4.1

-0.42.357.7Apr 2, 2026
Updated 16d ago

Evaluation Results

MethodLinks
2026.04
10100127.29921.4
2026.04
10100513.43326.5
2026.04
1010046.548-
2026.04
1010043.007-
2026.04
880144.97122.125
2026.04
660135.19615.1
2026.04
11014.159-
2026.04
11013.06-
2026.04
0027.541-
2026.04
0072.22-
2026.04
00277.07416.2
2026.04
0013.013-