Global MMLU
Global multilingual MMLU (SmolLM3 and Gemma 3n cards)
knowledge and reasoning
Score in %
Definition
- Version
- not in the notes
- Category
- knowledge and reasoning
- Maintainer
- not in the notes
- Status
- not in the notes
- Home page
- not in the notes
Sources
- HuggingFaceTB/SmolLM3-3B , Hugging Face huggingface.co https://huggingface.co/HuggingFaceTB/SmolLM3-3B
- Gemma 3n model card | Google AI for Developers ai.google.dev https://ai.google.dev/gemma/docs/gemma-3n/model_card
Scores at a Glance
Score in %
Higher is better for this unit.
1
Gemma 3n E4BGoogle DeepMindITGemma 3n E2BGoogle DeepMindIT
SmolLM3-3BHugging FaceCard 5-model comparison table; mode label not extracted
Gemma 3n E4BGoogle DeepMindIT60.3 %
vendor-reported2Gemma 3n E2BGoogle DeepMindIT55.1 %
vendor-reported3SmolLM3-3BHugging FaceCard 5-model comparison table; mode label not extracted53.5 %
vendor-reported0255075100
Where this comes from
Each bar is one model, with its most independent recorded score. Hatched bars are vendor-reported; the word beside each bar says who ran it. Settings and dates differ between rows; the Scores table below has every result.
Scores
| Model | Score | Who ran it | Setting | View or population | Date | Source |
|---|---|---|---|---|---|---|
| Gemma 3n E4B | 60.3 % | vendor-reported | IT | – | – | ai.google.dev |
| Gemma 3n E2B | 55.1 % | vendor-reported | IT | – | – | ai.google.dev |
| SmolLM3-3B | 53.5 % | vendor-reported | Card 5-model comparison table; mode label not extracted | – | – | huggingface.co |
Where this comes from
Scores from different settings, harnesses or dates are not directly comparable; the row says which. Sorted by score, highest first, as a reading aid, not a ranking.
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Global MMLU", https://aurion.semai.id/benchmarks/global-mmlu/, records of October 2026.