Grip on LLMs - Dutch tinyMMLU: leaderboard
Metric: Accuracy (%; GP-IRT estimate from 100 tinyBenchmarks items, Dutch machine translation). Source: arxiv.org. 31 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o | 70 |
| 2 | Mistral Large 3 | 69 |
| 3 | GPT-5 | 68 |
| 4 | Mistral Medium 3 | 66 |
| 5 | GPT-5 Mini | 65 |
| 6 | Gemma 3 27B (IT) | 60 |
| 7 | Qwen 3 32B | 60 |
| 8 | DeepSeek R1 Distill Qwen 32B | 57 |
| 9 | GPT-4o Mini | 56 |
| 10 | Gemma 3 12B (IT) | 54 |
| 11 | GPT-5 Nano | 54 |
| 12 | aya-expanse-32B | 54 |
| 13 | Mistral Small 3 | 52 |
| 14 | Qwen 3 8B | 50 |
| 15 | GPT-OSS-20B | 47 |
Interactive version: theaggregate.ai/benchmark?slug=grip-on-llms-dutch-tinymmlu · How It Works · Data refreshed daily, snapshot 2026-09-19.