GPQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
GPQA
91.5Accuracy
46
GPQA Diamond
87.35Accuracy
45
GPQA
95.2Accuracy (%)
44
GPQA
78.9Accuracy
44
GPQA
54.67Accuracy (GPQA)
43
GPQA
70.2pass@1
43
GPQA
0.382Accuracy
42
GPQA Diamond
98.74Accuracy
40
GPQA (80% test)
54.22Accuracy
39
GPQA Diamond
66.65Pass@1
39
GPQA
56.5Accuracy
38
GPQA
65.7pass@1
38
GPQA
63.6Accuracy
37
GPQA 0-shot
33.03Score
37
GPQA-Diamond Science N=198
41.4Accuracy
36
GPQA Diamond
58.08Accuracy
36
GPQA Diamond
85.5GPQA Diamond Reasoning Score
36
GPQA Diamond
1Total Inference Runtime (s)
36
GPQA
93.4pass@1
35
GPQA
69.16Score
35
GPQA
53.5EM
34
GPQA diamond
75.18Avg@8 Accuracy
34
GPQA
44Accuracy@1 (GPQA)
33
GPQA D
69.9Accuracy
33
GPQA Diamond
40Score
33