▸ multilingual niche
What is the best LLM
for French?
A French winner is published only when both independent per-language sources match callable models. This is a cross-source claim: Global-MMLU-Lite and the AI Language Proficiency Monitor must both report the model in French.
▸ evidence used
Global-MMLU-Lite Frenchindependent multilingual knowledge result
Language Proficiency Monitor FrenchARC + MGSM + MMLU accuracy panel
▸ what this rank does not claim
This score tests multilingual knowledge and reasoning in French. It does not prove tone, dialect quality, translation quality, or legal accuracy.
Models missing either required source are omitted rather than assigned a flattering score from incomplete evidence.