mmmlu@parity

MMMLU (Multilingual MMLU) parity validation subset with 10 tasks per language across 15 languages (150 tasks total). Evaluates language models’ subject knowledge and reasoning across multiple languages using multiple-choice questions covering 57 academic subjects.

← Back to Registry

Run this task

CLI:

inspect eval inspect_harbor/mmmlu_parity --model openai/gpt-5

Python:

from inspect_ai import eval
from inspect_harbor import mmmlu_parity

eval(mmmlu_parity(), model="openai/gpt-5")

Dataset information

Harbor registry	mmmlu@parity
Inspect task	mmmlu_parity
Version	parity
Samples	150
Paper	arxiv

See Task Parameters for the parameter set shared across all Harbor tasks.