Wikibench

Global-MMLU

reasoning

A comprehensive multilingual benchmark covering 42 languages that addresses cultural and linguistic biases in evaluation, with improved translation quality and culturally sensitive question subsets.

Leaderboard

Showing 5 of 5 results

  1. MiMo-V2.5-Pro
    83.6%
    i
  2. Gemma 3n E4B Instructed
    60.3%
    i
  3. Gemma 3n E4B Instructed LiteRT Preview
    60.3%
    i
  4. Gemma 3n E2B Instructed
    55.1%
    i
  5. Gemma 3n E2B Instructed LiteRT (Preview)
    55.1%
    i