Benchmark
Multilingual MMLU
MMLU-ProX is a comprehensive multilingual benchmark covering 29 typologically diverse languages, building upon MMLU-Pro. Each language version consists of…
- Modality
- text
- Categories
- reasoning, language, general
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 5
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Ministral 3 (14B Base 2512) | Mistral | 0.742 | 2025-12-04 | evidence |
| Ministral 3 (3B Base 2512) | Mistral | 0.652 | 2025-12-04 | evidence |
| Ministral 3 (8B Base 2512) | Mistral | 0.706 | 2025-12-04 | evidence |
| o3-mini | OpenAI | 0.807 | 2025-01-30 | evidence |
| Phi 4 Mini | Microsoft | 0.493 | 2025-02-01 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.