GPQA
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
GPQA
38.89Result Score
14
GPQA diamond
40.4Average@16 Accuracy
14
GPQA 0-shot (test)
162.95Throughput
14
GPQA
34.9Pass@1
14
GPQA Main
0.361Accuracy
14
GPQA
314TPF
14
GPQA Diamond
0.677PAQ Score
14
GPQA Bio
92.6Accuracy
14
GPQA
51.1Score
14
GPQA Diamond
94.9Accuracy
13
S-GPQA
57.3Accuracy (S-GPQA)
13
GPQA
52.02Accuracy
13
GPQA Diamond
54.6GPQA Diamond
13
GPQA
72Avg@8
13
GPQA-D
52.53Accuracy
13
GPQA D (val)
65.9Accuracy (%)
13
GPQA
35.04GPQA Accuracy
13
GPQA
9,032Avg Response Length
13
GPQA Diamond
70.1Accuracy
12
GPQA Diamond
8Samples Required
12
GPQA D
27.28Mean@16
12
GPQA Diamond
85.4General Reasoning Score
12
GPQA
50.6Overall Score
12
GPQA-D
5,318Thinking Tokens
12
GPQA Diamond
6,858Response Length (Tokens)
12