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

Science on GPQA

85.7Accuracy

GPT-5

-3.42819.71142.8565.989May 22, 2025Jun 29, 2025Aug 7, 2025Sep 15, 2025Oct 23, 2025Dec 1, 2025Jan 9, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
85.7
2026.01
84
2026.01
62.1
2025.05
54
2025.05
53
2025.05
52.5
2025.05
51
2025.05
51
2025.05
51
2025.05
50
2026.01
49.1
2025.05
49
2025.05
49
2025.05
49
2025.05
48.5
2025.05
48
2025.05
48
2025.05
47.5
2025.05
47
2025.05
47
2025.05
46.5
2026.01
46.4
2025.05
46
2025.05
45
2025.05
45
2026.01
44.4
2025.05
44.4
2025.05
43.4
2025.05
42.4
2026.01
42
2025.05
41.9
2026.01
41.06
2026.01
39.39
2026.01
39.39
2026.01
39.1
2026.01
38.64
2026.01
36.6
2026.01
34.8
2026.01
31.82
2026.01
30.6
2026.01
29.55
2026.01
29
2026.01
28.6
2026.01
28.03
2026.01
27.7
2026.01
26.8
2026.01
25.9
2026.01
24.6
2026.01
0