EuroEval Italian NLU - Sentipolc16: leaderboard
The "Sentipolc16" subtask of EuroEval Italian NLU. Italian-language EuroEval natural-language-understanding suite, separating NLU task performance from the broader all-task EuroEval aggregate.
Metric: Matthews correlation coefficient (%). Source: euroeval.com. Saturation forecast: Estimated already saturated. 435 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 75.6 | #66 |
| 2 | GPT-5 | 73.43 | #98 |
| 3 | Gemma 4 31B (IT) | 72.07 | #176 |
| 4 | Gemini 3 Flash (Preview) (Thinking) | 71.82 | #82 (Gemini 3 Flash (Preview)) |
| 5 | Claude Sonnet 4.5 (Non-reasoning) | 71.36 | #142 (Claude Sonnet 4.5) |
| 6 | GPT-5 (2025-08-07) (High) | 71.22 | |
| 7 | Claude 3.7 Sonnet (20250219) (Thinking) | 71.05 | #201 (Claude 3.7 Sonnet (20250219)) |
| 8 | O3 (2025-04-16) | 71 | #121 |
| 9 | Claude Sonnet 4.5 (Thinking) | 70.89 | #142 (Claude Sonnet 4.5) |
| 10 | Claude 3.5 Sonnet (20241022) | 70.47 | #291 |
| 11 | O4 Mini (2025-04-16) | 70.34 | #179 |
| 12 | Gemini 3 Flash (Preview) (Non-reasoning) | 69.78 | #82 (Gemini 3 Flash (Preview)) |
| 13 | Gemini 3.1 Flash Lite (Preview) | 69.24 | #194 |
| 14 | Gemini 2.5 Pro | 68.7 | #152 |
| 15 | GPT-5.4 Mini (2026-03-17) (High) | 68.44 | #169 (GPT-5.4 Mini (2026-03-17)) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-italian-nlu-sentipolc16 · How It Works · Data refreshed daily, snapshot 2026-10-20.