EuroEval Faroese NLU - FONE: leaderboard
EuroEval's Faroese natural-language-understanding score — aggregate over the EuroEval Faroese understanding tasks (reading comprehension, linguistic acceptability, NER), from the EuroEval European language model benchmark suite.
Metric: Micro F1 excluding MISC (%). Source: euroeval.com. Saturation forecast: Around January 2027. 401 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 84.9 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 2 | Gemma 4 31B (IT) | 84.74 | #176 |
| 3 | Gemini 3 Flash (Preview) (Thinking) | 83.89 | #82 (Gemini 3 Flash (Preview)) |
| 4 | Claude Sonnet 4.5 (Thinking) | 83.21 | #142 (Claude Sonnet 4.5) |
| 5 | Gemma 4 26B A4B (IT) | 82.95 | #297 |
| 6 | Qwen 3.5 35B A3B Base | 81.82 | #310 |
| 7 | Llama 3.1 405B Instruct FP8 | 81.57 | #389 |
| 8 | Gemini 3 Flash (Preview) (Non-reasoning) | 81.04 | #82 (Gemini 3 Flash (Preview)) |
| 9 | GPT-5.4 Mini (2026-03-17) (High) | 80.76 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 10 | Llama 3.1 70B | 80.08 | #592 |
| 11 | Mistral Small 3.1 | 79.69 | #610 |
| 12 | Voxtral-Small-24B-2507 | 79.51 | #581 |
| 13 | Claude Sonnet 4.5 (Non-reasoning) | 79.34 | #142 (Claude Sonnet 4.5) |
| 14 | Qwen 3 4B 2507 (Thinking) | 79.02 | #538 (Qwen 3 4B 2507) |
| 15 | Mistral Small 3.2 | 78.05 | #605 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-faroese-nlu-fone · How It Works · Data refreshed daily, snapshot 2026-10-20.