Multilingual Massive Multitask Language Understanding (MMMLU) Benchmark

Explore score history ↗

Multilingual Massive Multitask Language Understanding (MMMLU) is an AI benchmark in the Benchmark Radar catalog.

Released: 2024-09-13
Openness: unknown
Importer: Benchmark Saturation

Source provenance

Original evidence

Multilingual Massive Multitask Language Understanding (MMMLU) results and reported scores

Reported benchmark scores · Benchmark Saturation

Reported scores: 3

Model	Organization	Reported value	Reported	Evidence
GPT-4.5	—	85.1	—	evidence
GPT-4o	—	81.5	—	evidence
o3 mini	—	81.1	—	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.

Which sources cite Multilingual Massive Multitask Language Understanding (MMMLU)?

Multilingual Massive Multitask Language Understanding (MMMLU)