EuroEval Dutch Simplification - Duidelijke Taal: leaderboard
EuroEval's Dutch text-simplification task on the Duidelijke Taal dataset — models rewrite Dutch text in simpler language, scored with automatic overlap metrics, from the EuroEval European language model benchmark suite.
Metric: Score (%). Source: euroeval.com. Saturation forecast: Around August 2027. 183 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4.1 Mini | 58.12 | #355 |
| 2 | GPT-4.1 Nano | 57.78 | #731 |
| 3 | Qwen 3 235B A22B 2507 Instruct | 57.52 | #300 |
| 4 | GPT-OSS-20B (Low) | 56.94 | #518 (GPT-OSS-20B) |
| 5 | Llama 3.1 405B Instruct FP8 | 56.6 | #389 |
| 6 | EuroLLM-9B | 56.49 | #1254 |
| 7 | Gemma 4 26B A4B (IT) | 56.42 | #297 |
| 8 | Qwen 3 14B (Non-reasoning) | 56.34 | #537 (Qwen 3 14B) |
| 9 | gemma-3-27B-pt | 56.24 | #502 |
| 10 | GPT-OSS-120B | 55.82 | #341 |
| 11 | Qwen 3 235B A22B 2507 FP8 (Thinking) | 55.21 | #246 (Qwen 3 235B A22B 2507 FP8) |
| 12 | EuroLLM-22B-Instruct-2512 | 54.88 | #701 |
| 13 | GPT-5.2 | 54.77 | #110 |
| 14 | Qwen 3 32B (Non-reasoning) | 54.71 | #443 (Qwen 3 32B) |
| 15 | GPT-4.1 | 54.68 | #248 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-dutch-simplification-duidelijke-taal · How It Works · Data refreshed daily, snapshot 2026-10-20.