EuroEval Danish Knowledge - Danske Talemaader: leaderboard
Multiple-choice questions on the meaning of Danish idioms and proverbs (Danske Talemaader). One of the two datasets averaged into EuroEval's knowledge score for this language; few-shot by default, with some zero-shot and validation-split runs; Matthews correlation x100 as the primary metric (signed, chance 0), higher is better.
Metric: MCC (x100). Source: euroeval.com. Saturation forecast: Estimated already saturated. 369 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash (Preview) (Thinking) | 98.5 | #90 (Gemini 3 Flash (Preview)) |
| 2 | Gemini 2.5 Pro | 95.12 | #159 |
| 3 | Gemini 3 Flash (Preview) (Non-reasoning) | 95.08 | #90 (Gemini 3 Flash (Preview)) |
| 4 | O3 (2025-04-16) | 93.43 | #125 |
| 5 | Gemini 2.5 Pro (Preview 03-25) | 92.52 | #126 |
| 6 | Gemini 2.5 Flash (Thinking) | 92.5 | #255 (Gemini 2.5 Flash) |
| 7 | Claude Sonnet 4.6 | 92.46 | #94 |
| 8 | GPT-4.1 | 91.43 | #258 |
| 9 | GPT-5 (2025-08-07) (High) | 90.83 | |
| 10 | GPT-5.4 Mini (2026-03-17) (High) | 90.63 | #180 (GPT-5.4 Mini (2026-03-17)) |
| 11 | Grok 4.1 Fast (Reasoning) | 90.14 | #227 (Grok 4.1 Fast) |
| 12 | GPT-5 | 89.75 | #105 |
| 13 | Gemini 3 Pro (Preview) | 89.63 | #71 |
| 14 | Gemini 2.0 Flash (001) | 89.01 | #365 |
| 15 | Gemini 2.5 Flash (Preview 04-17) | 88.92 | #246 |
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-knowledge-danske-talemaader · How It Works · Data refreshed daily, snapshot 2026-11-05.