Science Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
57.3Accuracy (S-GPQA)
13
May 25, 2026
91.8Accuracy
13
May 22, 2026
44.1Accuracy
13
May 19, 2026
80.4Accuracy
13
Apr 21, 2026
65.9Accuracy (%)
13
Apr 14, 2026
68.5Accuracy
12
Jun 5, 2026
95.8Accuracy
12
Jun 4, 2026
28.46Avg@k
12
May 13, 2026
35.66Avg@k
12
May 13, 2026
38.22Accuracy
12
Apr 7, 2026
81.19Verifiability Score
12
Feb 26, 2026
49.45Reusability Score
12
Feb 26, 2026
92Accuracy
12
Feb 26, 2026
100Performance Score (Act)
11
Jun 30, 2026
84.34Accuracy
11
Jun 19, 2026
89ARC Score
11
Feb 26, 2026
91Accuracy
11
Feb 26, 2026
92Accuracy
11
Feb 26, 2026
34.6Score
11
Mar 25, 2026
71.6Accuracy
11
Feb 26, 2026
98.53Score
10
Jul 7, 2026
77.31Score
10
May 29, 2026
68.56Average@4
10
May 27, 2026
24.6Mean Accuracy
10
Feb 27, 2026
35.67Score
10
Feb 26, 2026