GPQA
Benchmarks
TaskDatasetSOTA resultTrendResults
GPQA
2Acceptance Length
6
GPQA
5.2Acceptance Length
6
GPQA Diamond
37.18Pass@1 Accuracy
6
GPQA
56.7Selection Accuracy
6
GPQA Diamond
55.6Avg@8
6
GPQA Diamond
86.3Throughput (Req/s)
6
GPQA
38.26GPQA Score
6
GPQA
92.08TPS
6
GPQA
193.62TPS
6
GPQA
41.4GPQA Accuracy
6
GPQA Diamond
39.6Avg@64
6
GPQA Diamond
20.7Genuine Followup Rate
6
GPQA Diamond OOD
71.5Sav.
6
GPQA-D
46.46Acc@8
6
GPQA
80.8Accuracy
6
GPQA Diamond
86.6Score
6
GPQA
-0.55FPR
6
GPQA Diamond
20.2Pass@1
6
GPQA Diamond
82.2avg@3
6
GPQA Diamond (test)
42.42Score
6
GPQA MC
37.8Pass@1
6
GPQA diamond
78.29Pass@10
6
GPQA main
76.7Pass@10
6
GPQA diamond
83.84Pass@20
6
GPQA main
83.26Pass@20
6