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

Scientific Question Answering on GPQA (Pass@8)

79.29Pass@8

MOTAB

52.5159.462566.41573.3675May 19, 2026
Updated 14d ago

Evaluation Results

MethodLinks
2026.05
79.29
2026.05
77.27
2026.05
76.77
2026.05
73.74
2026.05
73.74
2026.05
73.74
2026.05
71.21
2026.05
70.71
2026.05
70.71
2026.05
69.7
2026.05
68.18
2026.05
67.68
2026.05
66.67
2026.05
64.65
2026.05
63.13
2026.05
62.63
2026.05
61.62
2026.05
60.1
2026.05
58.59
2026.05
56.06
2026.05
53.54