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

Reasoning on GPQA-D (avg@k, Token usage)

77.46Avg@k

Type IV (STOP)

30.514442.702254.8967.0778Apr 17, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.04
77.46143.4-48.26
2026.04
68.43145.9-47.36
2026.04
66.41151.8-45.24
2026.04
65.55277.2-
2026.04
63.32193.5-70.35
2026.04
58.78196.6-69.86
2026.04
57.07201-69.2
2026.04
56.87652.6-
2026.04
55.75165.9-62.61
2026.04
48.42179.4-67.43
2026.04
47.09173.6-60.89
2026.04
46.02166.7-62.43
2026.04
45.95443.8-
2026.04
34.98180.4-67.25
2026.04
33.08550.9-
2026.04
32.32187.1-66.03