NorskGPT-Mistral-7B: benchmark results

Provider: Bineric. Released 2024-01-02. Access: Open.

Aggregate Rating 1410 ± 37, rank #5600 of 10916 rated models, from 21 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
EuroEval Swedish Common Sense Reasoning43.72Common Sense Reasoning Average Score (%)59.4
EuroEval Danish Knowledge62.53Knowledge Average Score (%)57.6
EuroEval Swedish NLU - SUC358.4Micro F1 excluding MISC (%)57.5
EuroEval Danish Common Sense Reasoning43.42Common Sense Reasoning Average Score (%)56.7
EuroEval Norwegian Common Sense Reasoning61.44Common Sense Reasoning Average Score (%)56.5
EuroEval English Knowledge76.8Knowledge Average Score (%)56.1
EuroEval Danish Knowledge - Danish Citizen Tests62.53MCC (x100)53.9
EuroEval Danish NLU - Dansk50.76Micro F1 excluding MISC (%)52
EuroEval Norwegian Knowledge - NRK Quiz QA28.43MCC (x100)50.1
EuroEval Swedish NLU - Swerec74.3Matthews correlation coefficient (%)47.9
EuroEval Norwegian Knowledge13.63Knowledge Average Score (%)46.4
EuroEval Swedish Knowledge35.01Knowledge Average Score (%)46

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