EuroEval Icelandic NLU - NQII: leaderboard
The "NQII" subtask of EuroEval Icelandic NLU. Icelandic-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Character F1 (%). Source: euroeval.com. Saturation forecast: Not forecast. 384 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen3.5-4B-Base | 65.37 | #798 |
| 2 | Llama 3.1 70B | 63.64 | #592 |
| 3 | Mistral Small 3.1 | 63.53 | #610 |
| 4 | Llama 3.1 405B | 62.33 | #466 |
| 5 | Ministral-3-14B-Reasoning-2512 | 62.18 | #540 |
| 6 | Gemma 2 27B | 62.06 | #806 |
| 7 | gemma-3-27B-pt | 62.01 | #502 |
| 8 | Mistral Small 3.2 | 61.42 | #605 |
| 9 | gemma-3-12B-pt | 61.39 | #686 |
| 10 | Magistral Small 1.2 | 60.76 | #570 |
| 11 | Llama-3.1-Tulu-3-8B-SFT | 60.51 | #985 |
| 12 | Gemma 3n E4B | 60.23 | #878 |
| 13 | Qwen 3 14B (Non-reasoning) | 59.53 | #537 (Qwen 3 14B) |
| 14 | Llama 2 70B Base | 59.39 | #1025 |
| 15 | Voxtral-Small-24B-2507 | 58.98 | #581 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-icelandic-nlu-nqii · How It Works · Data refreshed daily, snapshot 2026-10-20.