Global-MMLU-Lite

reasoning

A lightweight version of Global MMLU benchmark that evaluates language models across multiple languages while addressing cultural and linguistic biases in multilingual evaluation.

Leaderboard

Showing 14 of 14 results

Gemini 2.5 Pro Preview 06-05

89.2%

i
Gemini 2.5 Pro

88.6%

i
Gemini 2.5 Flash

88.4%

i
Gemini 2.5 Flash-Lite

81.1%

i
Gemini 2.0 Flash-Lite

78.2%

i
Gemma 3 27B

75.1%

i
Gemma 3 12B

69.5%

i
Gemini Diffusion

69.1%

i
Gemma 3n E4B Instructed

64.5%

i
Gemma 3n E4B Instructed LiteRT Preview

64.5%

i
Gemma 3n E2B Instructed

59.0%

i
Gemma 3n E2B Instructed LiteRT (Preview)

59.0%

i
Gemma 3 4B

54.5%

i
Gemma 3 1B

34.2%

i