Multilingual Mathematical Reasoning
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
93.4Accuracy
109
May 6, 2026
93.6Accuracy
64
Mar 4, 2026
94.9Accuracy
64
Jun 30, 2026
69.6Accuracy (ru)
35
Feb 26, 2026
75Accuracy (English)
33
Feb 26, 2026
20.3Accuracy (Ar)
30
May 22, 2026
41.6Accuracy
25
May 19, 2026
28.6Average Score (All)
22
May 22, 2026
30.1Vietnamese Accuracy
22
May 22, 2026
26.3Accuracy (Ar)
22
May 22, 2026
6.08Speedup (de)
20
Feb 26, 2026
89Accuracy
12
Jun 16, 2026
89.4Accuracy
12
Jun 16, 2026
88.4Accuracy
12
Jun 16, 2026
85.9Accuracy
12
Jun 16, 2026
83.6Accuracy
12
Jun 16, 2026
90.4Accuracy
12
Jun 16, 2026
88.27Accuracy
11
Jun 26, 2026
79.9FR SLC&Acc
8
Jul 2, 2026
41.2FR Accuracy and SLC
8
Jul 2, 2026
47.6Accuracy (En)
7
Feb 26, 2026
84.89MT-MATH100 Score
7
Feb 26, 2026
72.5Accuracy
6
Feb 26, 2026
87.73Native CoT Avg Score
5
Jun 16, 2026