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

Graduate-Level Reasoning on GPQA

78.9Accuracy

M2CL

17.85233.70149.5565.399Jul 15, 2024Oct 31, 2024Feb 17, 2025Jun 5, 2025Sep 22, 2025Jan 8, 2026Apr 27, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
78.9
2026.04
73.1
2026.04
73
2026.04
72.2
2026.02
66.2
2026.02
66.1
2025.12
61.51
2025.12
61.21
2025.12
60.5
2025.12
60.12
2026.04
54.04
2026.02
52.4
2026.02
51.6
2026.04
51.01
2026.04
50.5
2026.04
49.49
2026.04
49.49
2026.04
48.99
2026.04
47.98
2026.04
46.97
2026.02
46.2
2026.04
45.96
2026.02
45.9
2026.02
43.4
2026.04
42.93
2026.04
41.92
2026.04
41.41
2026.02
39.4
2026.02
38.7
2026.02
36.4
2026.02
35.6
2026.02
35.4
2026.02
34.9
2024.07
34.3
2026.02
32.8
2026.02
31.9
2024.07
30.8
2024.07
30.8
2026.02
30.8
2026.02
30.3
2024.07
29.2
2026.02
28.7
2026.02
21.2
2026.02
20.2