EuroEval Romanian NLU - MultiWikiQA RO — leaderboard
Metric: Reading comprehension Score (%). Source: euroeval.com. 207 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | SOLAR-10.7B-v1.0 | 77.57 |
| 2 | Bielik-11B-v2.3-Instruct | 76.93 |
| 3 | Mistral Small 3.2 | 76.89 |
| 4 | gemma-3-27B-pt | 76.12 |
| 5 | Llama 3.1 70B | 75.64 |
| 6 | Qwen 3 14B (Non-reasoning) | 75.44 |
| 7 | Ministral 3 14B | 74.56 |
| 8 | gemma-3-12B-pt | 73.8 |
| 9 | Mistral Small 3.1 | 73.72 |
| 10 | Ministral-3-14B-Reasoning-2512 | 72.98 |
| 11 | Llama 3.1 8B | 72.78 |
| 12 | Ministral 3 3B | 72.31 |
| 13 | gemma-3-4B-pt Base | 71.92 |
| 14 | Ministral-3-3B-Reasoning-2512 | 71.92 |
| 15 | Olmo 3.1 32B Instruct | 71.88 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-romanian-nlu-multiwikiqa-ro · How the rankings work · Data refreshed daily, snapshot 2026-07-22.