EuroEval Italian NLU - SQuAD IT: leaderboard
Metric: Reading comprehension Score (%). Source: euroeval.com. 429 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Ministral 3 14B | 78.66 |
| 2 | Ministral-3-14B-Reasoning-2512 | 78.42 |
| 3 | SOLAR-10.7B-v1.0 | 77.48 |
| 4 | Magistral Small 1.2 | 76.71 |
| 5 | Ministral 3 8B | 76.39 |
| 6 | Mistral Small 3.2 | 76.1 |
| 7 | Llama 3.1 405B | 75.85 |
| 8 | Ministral-3-14B-Instruct-2512 | 75.14 |
| 9 | Claude Sonnet 4.5 (Non-reasoning) | 75.1 |
| 10 | Gemma 2 27B | 74.8 |
| 11 | Mistral Small 3.1 | 74.76 |
| 12 | Yi 1.5 34B | 74.69 |
| 13 | Llama 3.1 70B | 74.66 |
| 14 | Tiny Aya Global | 74.62 |
| 15 | Llama 3 70B | 74.61 |
Interactive version: theaggregate.ai/benchmark?slug=euroeval-italian-nlu-squad-it · How It Works · Data refreshed daily, snapshot 2026-09-05.