Llama 2 7B Chat Norwegian: benchmark results

Norwegian chat-tuned Llama 2 7B checkpoint. Provider: Meta. Access: Open.

Aggregate Rating 1220 ± 35, rank #9475 of 10916 rated models, from 50 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
EuroEval Icelandic NLU - ScaLA IS0.07Matthews correlation coefficient (%)25.1
EuroEval Norwegian Knowledge - Idioms NO0.05MCC (x100)20.8
EuroEval Dutch NLU - DBRD49.2Matthews correlation coefficient (%)18.9
EuroEval Dutch NLU - ScaLA NL2.52Matthews correlation coefficient (%)17.8
EuroEval Dutch Knowledge5.41Knowledge Average Score (%)16.3
EuroEval Dutch NLU - CoNLL NL35.49Micro F1 excluding MISC (%)16.3
EuroEval Norwegian NLU - Norec12.22Matthews correlation coefficient (%)15.5
EuroEval Dutch NLU29.07NLU Average Score (%)14.6
EuroEval Norwegian Knowledge - NRK Quiz QA2.79MCC (x100)14.6
EuroEval Norwegian Knowledge1.42Knowledge Average Score (%)14.2
EuroEval Icelandic Knowledge0.39Knowledge Average Score (%)13.9
EuroEval Norwegian6.73Average Score (%)13.7

Interactive version: theaggregate.ai/model?slug=llama-2-7b-chat-norwegian · How It Works · Data refreshed daily, snapshot 2026-11-08.