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

Task Success on MemGUI-Bench 1.0 (test)

66.7Pass@1 L1

Claude-Sonnet-4.6

-2.66815.34133.3551.359May 28, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.05
66.770.451.96377.877.870.475.3
2026.05
48.13722.235.874.151.944.456.8
2026.05
3730.833.333.770.459.374.167.9
2026.05
25.922.222.223.43733.340.737
2026.05
22.214.818.518.529.622.222.224.7
2026.05
22.218.518.519.725.922.225.924.7
2026.05
18.57.43.79.940.714.814.823.4
2026.05
11.107.46.225.97.418.517.3
2026.05
7.47.404.914.814.87.412.3
2026.05
3.77.414.88.618.511.125.918.5
2026.05
00000000