EuroEval Faroese NLU - FoQA — leaderboard

Metric: Reading comprehension Score (%). Source: euroeval.com. 393 models tracked.

Top models

#ModelScore
1Ministral 3 14B79.94
2Mistral Small 3.279.92
3Llama 3.1 405B Instruct FP879.74
4Magistral Small79.56
5Mistral Small 3.179.48
6Llama 3.1 70B79.34
7Ministral-3-14B-Reasoning-251279.31
8Ministral-3-14B-Instruct-251278.86
9Gemma 2 9B (IT)78.56
10gemma-3-12B-pt77.27
11gemma-3-27B-pt77.23
12GPT-5.4 Mini (Non-reasoning)77.19
13Gemini 2.5 Flash Lite (Non-reasoning)76.6
14GPT-5 Mini (Minimal)76.55
15GPT-4.176.47

Interactive version: theaggregate.ai/benchmark?slug=euroeval-faroese-nlu-foqa · How the rankings work · Data refreshed daily, snapshot 2026-07-22.