EuroEval Danish Summarization - Nordjylland News: leaderboard
EuroEval's Danish summarization task on the Nordjylland News dataset — models write abstractive summaries of Danish news articles, scored with automatic text-overlap metrics, from the EuroEval European language model benchmark suite.
Metric: chrF3++ (%). Source: euroeval.com. Saturation forecast: Around 2034. 160 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemma 4 31B (IT) | 38.54 | #176 |
| 2 | Qwen 3.5 35B A3B FP8 | 38.49 | #362 |
| 3 | Gemma 4 26B A4B (IT) | 38.1 | #297 |
| 4 | Mistral Small 3.1 | 37.93 | #610 |
| 5 | Gemma 3 12B (IT) | 37.81 | #664 |
| 6 | Gemini 3 Pro (Preview) | 37.71 | #66 |
| 7 | Qwen 3.5 35B A3B Base | 37.63 | #310 |
| 8 | Llama 3.3 70B Instruct | 37.58 | #534 |
| 9 | Gemma 3 27B (IT) | 37.47 | #523 |
| 10 | GPT-4.1 | 37.45 | #248 |
| 11 | EuroLLM-22B-Instruct-2512 | 37.44 | #701 |
| 12 | gemma-4-E4B-it | 37.41 | #566 |
| 13 | Mistral Small 3.2 | 37.34 | #605 |
| 14 | Gemma 4 E2B (IT) | 37.24 | #781 |
| 15 | Llama 3.1 8B Instruct | 37.16 | #1035 |
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-summarization-nordjylland-news · How It Works · Data refreshed daily, snapshot 2026-10-20.