Science Question Answering
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
94.1Accuracy
16
May 26, 2026
91.5Accuracy
16
Feb 26, 2026
92.11ARC Easy Accuracy
16
Feb 26, 2026
18,851.29Avg Context Length (tokens)
16
Feb 26, 2026
65Task Accuracy
15
Jul 9, 2026
74.7SQA^I Score
15
Mar 24, 2026
79.5Accuracy
15
Mar 6, 2026
96.93Correct Prediction Rate
15
Feb 26, 2026
90Accuracy
15
Feb 26, 2026
91Accuracy
15
Feb 26, 2026
95.19Accuracy
14
Jun 16, 2026
47.98Accuracy
14
Jun 4, 2026
72.3Accuracy
14
May 26, 2026
58.33Accuracy (avg@k)
14
May 12, 2026
0.88ECE
14
Apr 21, 2026
84.98Accuracy
14
Mar 18, 2026
73.5Accuracy
14
Mar 6, 2026
162.95Throughput
14
Mar 20, 2026
98.76Exact Match
14
Mar 25, 2026
0.361Accuracy
14
Feb 26, 2026
25.8Accuracy
13
Jul 7, 2026
67.9Accuracy
13
May 27, 2026
70.1SQA-I
13
May 27, 2026
97.1Accuracy
13
May 25, 2026
70Accuracy
13
May 25, 2026