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

Science Reasoning Accuracy on GPQA

54.67Accuracy (GPQA)

EAPO

-2.186812.574127.33542.0959Aug 5, 2025Sep 28, 2025Nov 22, 2025Jan 16, 2026Mar 11, 2026May 5, 2026Jun 29, 2026
Updated 23d ago

Evaluation Results

MethodLinks
2026.06
54.67
2026.06
52.27
2026.06
52.27
2026.06
50.51
2026.06
50.44
2026.06
43.18
2026.06
41.98
2026.06
39.67
2026.06
39.39
2026.06
39.27
2026.06
38.76
2026.06
34.34
2025.10
30.8
2025.10
30.3
2025.10
30.3
2025.10
29.3
2025.10
27.8
2025.10
26.8
2025.10
25.8
2025.10
25.3
2025.10
22.7
2025.10
17.7
2025.10
16.2
2025.10
15.2
2025.10
11.1
2025.08
4.35
2025.08
4.35
2025.08
4.17
2025.08
3.99
2025.08
3.44
2025.08
3.1
2025.08
3.08
2025.10
3
2025.08
2.98
2025.08
2.54
2025.08
2.54
2025.08
2.53
2025.08
2.36
2025.08
2.17
2025.08
2.17
2025.08
1.45
2025.08
1.27
2025.10
0