Arithmetic Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
100Accuracy
324
Jun 19, 2026
97.1Accuracy
272
May 27, 2026
97.35Accuracy
203
Jun 26, 2026
99Accuracy
149
May 27, 2026
99.5Accuracy
136
Jun 26, 2026
100Accuracy
106
Apr 15, 2026
96.01Accuracy
87
May 27, 2026
98.16Accuracy
84
Jun 26, 2026
98.8Accuracy
73
May 27, 2026
74.63Accuracy
72
Jun 26, 2026
93.74Accuracy
69
Jun 4, 2026
77.1Accuracy
57
May 27, 2026
93.7Accuracy (Overall)
54
Feb 26, 2026
53.4Accuracy
50
Feb 26, 2026
92.9Average Score
46
Apr 24, 2026
71Accuracy
39
May 11, 2026
62.2AQuA Accuracy
31
Mar 4, 2026
95.5Accuracy
30
Feb 26, 2026
90Success Rate
28
May 26, 2026
84.3Accuracy
26
May 27, 2026
63.1Accuracy
25
Jun 9, 2026
99.67MultiArith Accuracy
24
Apr 21, 2026
95.94Accuracy
23
Jun 4, 2026
96.71Accuracy
22
Jun 26, 2026
90.7Accuracy
18
May 28, 2026