Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Multilingual Benchmark

Benchmarks

Task NameDataset NameSOTA ResultTrend
CombinedMultilingual Benchmark
IA Score8.83
34
ReasoningMultilingual Benchmark
IA Score6.04
17
StyleMultilingual Benchmark
IA8.82
17
TranslateMultilingual Benchmark
IA6.34
17
RearrangeMultilingual Benchmark
IA Score7.37
17
DeleteMultilingual Benchmark
IA8.6
17
ReplaceMultilingual Benchmark
IA8.39
17
AddMultilingual Benchmark
IA9.59
17
Multilingual NLPMultilingual Benchmark Average across languages (test)
Average Score89.15
8
Text RenderingMultilingual Benchmark English (test)
Character Precision99.68
7
Text RenderingMultilingual Benchmark Chinese (test)
Character Precision (Chars Pre)93.44
5
Showing 11 of 11 rows