Multilingual Understanding
Benchmarks
Dataset NameSOTA methodMetricTrendResultsLast Updated
81.5Score
32
May 27, 2026
1.014Performance
13
Jun 16, 2026
1.155Normalized Performance
13
Jun 16, 2026
1.235Normalized Performance
13
Jun 16, 2026
49.47Accuracy
12
Mar 16, 2026
47.78Accuracy
12
Mar 16, 2026
82Score
9
Jul 2, 2026
91.8MMMLU Score
9
Jul 2, 2026
95.6Score (Global PIQA)
9
Jul 2, 2026
57.8Accuracy
5
Jun 8, 2026
66.26CES Score
5
Feb 26, 2026
52.11CES Performance
5
Feb 26, 2026
98.47Accuracy
1
Jun 5, 2026