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

Science on SciBench (Accuracy)

28.3Accuracy (SciBench)

SC

13.94817.67421.425.126May 22, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.05
28.3
2025.05
28.3
2025.05
28.1
2025.05
28.1
2025.05
27.9
2025.05
27.9
2025.05
26.9
2025.05
26.9
2025.05
26.9
2025.05
26.7
2025.05
26.5
2025.05
26.3
2025.05
25.9
2025.05
25.7
2025.05
25.5
2025.05
25.3
2025.05
25.1
2025.05
24.7
2025.05
24.7
2025.05
24.3
2025.05
23.7
2025.05
23.5
2025.05
21.9
2025.05
14.5