Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Language Identification research benchmarks – Page 2 · SOTA2 Research

Language Identification

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

UDHR n=10,283 (test)

0.005FPR

7

Feb 26, 2026

FastSpell n=6,809 (excluding Nynorsk)

0.15FPR

7

Feb 26, 2026

99.71In-domain Acc

7

Feb 26, 2026

99.67In-domain Accuracy

7

Feb 26, 2026

90.33Accuracy (hi)

6

May 14, 2026

CommonVoice & Fleurs

100Accuracy (eng)

6

Feb 26, 2026

98.7Accuracy

6

Mar 12, 2026

CommonVoice (test)

98.7Accuracy

6

Mar 12, 2026

GlotLID-C 2099 languages (test)

ConLID-S+LID_{CE}

98.68F1 Score

5

Mar 11, 2026

FLORES-200 CLD3 label set 77 languages (test)

99.9F1 Score

5

Feb 26, 2026

UDHR CLD3 label set 80 languages (test)

0.998F1 Score

5

Feb 26, 2026

HPLT-LID 3.0 (manual inspection)

0.001FPR

5

Feb 26, 2026

89.68Macro F1

5

Feb 26, 2026

90.02Macro F1

5

Feb 26, 2026

70.38Macro F1 Score

5

Feb 26, 2026

93.54Macro F1

5

Feb 26, 2026

94.52Macro F1

5

Feb 26, 2026

95.76Macro F1

5

Feb 26, 2026

97.83Macro-F1

5

Feb 26, 2026

Voxforge (test)

99.8Accuracy

5

Feb 26, 2026

EU dataset White Noise

91.2Accuracy

5

Feb 26, 2026

EU dataset No Noise

96.7Accuracy

5

Feb 26, 2026

Indian Language dataset

0.987Accuracy

5

Feb 26, 2026

Fleurs Nigerian languages (test)

100F1 Score (yo)

4

May 19, 2026

98.61F1 Score

4

Mar 11, 2026

Page 2 of 4

Popular datasets

FLORES UDHR FLEURS ML-SUPERB GlotLID-C

Follow for updates

Join SOTA2 Contact