EuroEval Danish NLU - Angry Tweets: leaderboard
The "Angry Tweets" subtask of EuroEval Danish NLU. Danish-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Matthews correlation coefficient (%). Source: euroeval.com. Saturation forecast: Estimated already saturated. 436 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 (Thinking) | 64.26 | #142 (Claude Sonnet 4.5) |
| 2 | Gemini 2.5 Flash (Preview 04-17) | 63.76 | #237 |
| 3 | Gemma 4 31B (IT) | 63.59 | #176 |
| 4 | Qwen 3.5 27B FP8 | 63.48 | #164 |
| 5 | Claude Sonnet 4.5 (Non-reasoning) | 62.64 | #142 (Claude Sonnet 4.5) |
| 6 | O3 Mini (2025-01-31) | 62.16 | #240 |
| 7 | Gemini 3 Flash (Preview) (Thinking) | 61.73 | #82 (Gemini 3 Flash (Preview)) |
| 8 | GLM-4.7 FP8 | 61.64 | #209 |
| 9 | Llama 3.1 405B Instruct FP8 | 60.84 | #389 |
| 10 | Qwen 3.5 35B A3B FP8 | 60.55 | #362 |
| 11 | Gemini 3 Pro (Preview) | 60.17 | #66 |
| 12 | OLMo 3.1 32B (Thinking) | 59.97 | #632 (OLMo 3.1 32B) |
| 13 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 59.27 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 14 | Gemini 1.5 Pro (002) | 59.25 | #349 |
| 15 | Mistral Small 3.1 | 59.08 | #610 |
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-angry-tweets · How It Works · Data refreshed daily, snapshot 2026-10-20.