EuroEval Danish NLU: leaderboard
Danish-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: NLU Average Score (%). Source: euroeval.com. Saturation forecast: Estimated already saturated. 454 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 70.44 | #66 |
| 2 | Gemma 4 31B (IT) | 70.18 | #176 |
| 3 | Gemini 3.1 Flash Lite (Preview) | 70.05 | #194 |
| 4 | GPT-5 | 69.22 | #98 |
| 5 | Claude Sonnet 4.5 (Thinking) | 69.17 | #142 (Claude Sonnet 4.5) |
| 6 | Gemini 3 Flash (Preview) (Thinking) | 69.16 | #82 (Gemini 3 Flash (Preview)) |
| 7 | Claude 3.7 Sonnet (20250219) (Thinking) | 69.11 | #201 (Claude 3.7 Sonnet (20250219)) |
| 8 | GLM-4.7 FP8 | 68.61 | #209 |
| 9 | GPT-5.4 Mini (2026-03-17) (High) | 68.33 | #169 (GPT-5.4 Mini (2026-03-17)) |
| 10 | Gemini 3 Flash (Preview) (Non-reasoning) | 68.23 | #82 (Gemini 3 Flash (Preview)) |
| 11 | GPT-5.2 | 68.03 | #110 |
| 12 | O3 (2025-04-16) | 67.86 | #121 |
| 13 | Claude Sonnet 4.6 | 67.41 | #90 |
| 14 | Qwen 3.5 27B FP8 | 67.33 | #164 |
| 15 | GPT-5 (2025-08-07) (High) | 67.15 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-danish-nlu · How It Works · Data refreshed daily, snapshot 2026-10-20.