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

Reasoning on GPQA (80% test)

54.22Accuracy

Self-Refine

36.675241.230145.78550.3399May 21, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.05
54.22
2025.05
53.6
2025.05
52.61
2025.05
52.41
2025.05
52.01
2025.05
51.81
2025.05
51.2
2025.05
50.6
2025.05
50.6
2025.05
50.6
2025.05
50.6
2025.05
50
2025.05
49.8
2025.05
49.4
2025.05
49.4
2025.05
48.43
2025.05
47.59
2025.05
47.17
2025.05
47.17
2025.05
47
2025.05
46.99
2025.05
46.99
2025.05
46.99
2025.05
46.99
2025.05
46.39
2025.05
46.39
2025.05
46.18
2025.05
45.78
2025.05
45.2
2025.05
43.98
2025.05
43.98
2025.05
43.57
2025.05
43.37
2025.05
43.37
2025.05
42.77
2025.05
42.57
2025.05
41.57
2025.05
40
2025.05
37.35