EuroEval Dutch Common Sense Reasoning: leaderboard
Metric: Common Sense Reasoning Average Score (%). Source: euroeval.com. 319 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Sonnet 4.5 (Non-reasoning) | 93.05 |
| 2 | Claude Sonnet 4.5 (Thinking) | 93.04 |
| 3 | Gemini 3 Pro (Preview) | 85.18 |
| 4 | Qwen 3 235B A22B 2507 Instruct | 84.68 |
| 5 | Claude Sonnet 4.6 | 83.48 |
| 6 | GPT-5 (High) | 83.46 |
| 7 | GPT-5 (Minimal) | 81.97 |
| 8 | Gemma 4 31B (IT) | 80.62 |
| 9 | Gemini 2.5 Pro | 80.57 |
| 10 | Mistral Small 3.1 | 80.14 |
| 11 | Llama 3.1 405B Instruct FP8 | 80.03 |
| 12 | GPT-5 | 79.64 |
| 13 | Qwen 3 Next 80B A3B Instruct | 79.09 |
| 14 | Mistral Small 3 | 79.04 |
| 15 | GPT-4.1 | 78.25 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-dutch-common-sense-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.