openai/mmmlu
Knowledge
Reasoning
MMMLU (Multilingual MMLU): OpenAI’s professional-human-translation of the MMLU test set into 14 languages for multilingual knowledge and reasoning evaluation.
Run this task
CLI:
inspect eval inspect_harbor/openai_mmmlu --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import openai_mmmlu
eval(openai_mmmlu(), model="openai/gpt-5")Dataset information
| Harbor registry | openai/mmmlu |
| Inspect task | openai_mmmlu |
| Latest digest | sha256:5db8efae92fcb2df5fb3c76647394410badcd08cec58a3cdfd3c602f7d9b38d1 |
| Samples | 150 |
| Paper | arxiv |
| Source | https://huggingface.co/datasets/openai/MMMLU |
See Task Parameters for the parameter set shared across all Harbor tasks.