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

Science Question Answering on GPQA (Accuracy and Delta Avg)

91.5Accuracy

CoT2-Meta

8.851230.308151.76573.2219Oct 7, 2025Nov 6, 2025Dec 6, 2025Jan 5, 2026Feb 4, 2026Mar 6, 2026Apr 6, 2026
Updated 11d ago

Evaluation Results

MethodLinks
2026.03
91.510.5
2026.03
90.414.5
2026.03
88.35.7
2026.03
85.13
2026.03
83.58.3
2026.03
82.4-
2026.03
79.64.8
2026.03
74.2-
2026.03
59.812.2
2026.03
53.46.4
2026.03
49.23.4
2026.03
44.7-
2025.10
42.63-
2025.10
40.4-
2025.10
39.51-
2025.10
38.84-
2025.10
38.17-
2026.04
36.01-
2025.10
35.27-
2025.10
33.48-
2025.10
33.03-
2025.10
31.92-
2026.04
31.81-
2025.10
31.47-
2025.10
30.36-
2025.10
30.36-
2025.10
29.91-
2025.10
29.69-
2025.10
28.35-
2025.10
27.46-
2025.10
27.23-
2025.10
27.01-
2026.04
26.12-
2025.10
25.45-
2026.04
24.72-
2025.10
23.88-
2026.04
23.57-
2026.04
23.34-
2026.04
23.04-
2026.04
22.27-
2026.04
17.89-
2026.04
16.98-
2026.04
16.13-
2026.04
15.79-
2026.04
15.32-
2026.04
12.03-