Title: European languages %E2%80%94 Benchmarks | LLM EU
Canonical: https://llmeu.com/sv/benchmarks/eu-languages/
Markdown mirror: https://llmeu.com/sv/benchmarks/eu-languages.md
Machine-readable: https://llmeu.com/llms.txt (index) - https://llmeu.com/llms-full.txt (full catalogue) - https://llmeu.com/llms-catalogue.json

# European languages

Modellkvalitet p%C3%A5 %C3%B6versatta europeiska spr%C3%A5kbenchmarks. EU21-raderna kommer fr%C3%A5n OpenGPT-X-utv%C3%A4rderingen som publicerades med Teuken 7B Instruct v0.6, som %C3%B6vers%C3%A4tter ARC, HellaSwag, TruthfulQA och MMLU till de 21 EU-spr%C3%A5k den t%C3%A4cker och rapporterar ett genomsnitt per modell. MMMLU-raden %C3%A4r Googles egen flerspr%C3%A5kiga MMLU-siffra f%C3%B6r Gemma 4.

The EU21 numbers are OpenGPT-X%E2%80%99s, taken from the Teuken 7B Instruct v0.6 model card and the accompanying evaluation preprint; they are machine-translated benchmarks, not native-language test sets. The MMMLU number is Google%E2%80%99s own. LLM EU has run no evaluation and has not verified either harness.