reasoning benchmark
Global-MMLU
A comprehensive multilingual benchmark covering 42 languages that addresses cultural and linguistic biases in evaluation, with improved translation quality and culturally sensitive question subsets.
Updated Aug 7, 2026
Published models5
Registry coverage5
MetricScore
EvidenceB
Global-MMLU leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | XI | 83.6% | 100.0% | 5 | C | |
| 2 | GO | 60.3% | 75.0% | 5 | C | |
| 3 | GO | 60.3% | 50.0% | 5 | C | |
| 4 | GO | 55.1% | 25.0% | 5 | C | |
| 5 | GO | 55.1% | 0.0% | 5 | C |
Score distribution
Top published rows on the benchmark's original scale.
Global-MMLU highlights
The top published results on this benchmark's own scale.
What is Global-MMLU?
Definition and scoring fields from the benchmark registry.
A comprehensive multilingual benchmark covering 42 languages that addresses cultural and linguistic biases in evaluation, with improved translation quality and culturally sensitive question subsets.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- Global-MMLU
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- global-mmlu|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about Global-MMLU.
Which model scores highest on Global-MMLU?
MiMo-V2.5-Pro is currently ranked first with 83.6%.
What does Global-MMLU measure?
A comprehensive multilingual benchmark covering 42 languages that addresses cultural and linguistic biases in evaluation, with improved translation quality and culturally sensitive question subsets.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
5 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.