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

Scientific Question Answering on GPQA (average inference time (seconds))

1.58Average Inference Time (s)

CoT

-3.503630.810765.12599.4393Sep 24, 2025
Updated 24d ago

Evaluation Results

MethodLinks
2025.09
1.58
2025.09
2.44
2025.09
3.92
2025.09
6
2025.09
6.37
2025.09
7.61
2025.09
7.63
2025.09
9.54
2025.09
11.45
2025.09
12.77
2025.09
13.24
2025.09
16.05
2025.09
17.04
2025.09
22.21
2025.09
23.11
2025.09
29.2
2025.09
33.78
2025.09
36.83
2025.09
38.08
2025.09
49.02
2025.09
49.77
2025.09
52.33
2025.09
68.17
2025.09
79.24
2025.09
79.33
2025.09
84.06
2025.09
85.67
2025.09
102.78
2025.09
104.25
2025.09
128.67