EuroEval Portuguese NLU - SST-2 PT: leaderboard
The "SST-2 PT" subtask of EuroEval Portuguese NLU. Portuguese-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. 444 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 3 Flash (Preview) (Thinking) | 92.35 | #82 (Gemini 3 Flash (Preview)) |
| 2 | Claude Sonnet 4.6 | 91.66 | #90 |
| 3 | Gemini 3 Pro (Preview) | 90.88 | #66 |
| 4 | Gemini 3.1 Flash Lite (Preview) | 89.25 | #194 |
| 5 | Ministral-3-14B-Reasoning-2512 | 89.1 | #540 |
| 6 | Gemma 4 31B (IT) | 88.97 | #176 |
| 7 | Claude Sonnet 4.5 (Thinking) | 88.86 | #142 (Claude Sonnet 4.5) |
| 8 | Gemma 4 26B A4B (IT) | 88.68 | #297 |
| 9 | Gemini 3 Flash (Preview) (Non-reasoning) | 88.38 | #82 (Gemini 3 Flash (Preview)) |
| 10 | GPT-5 (2025-08-07) (High) | 88.2 | |
| 11 | Bielik-11B-v2.3-Instruct | 87.85 | #693 |
| 12 | Ministral-3-14B-Instruct-2512 | 87.08 | #615 |
| 13 | gemma-4-E4B-it | 86.93 | #566 |
| 14 | GPT-5 | 86.82 | #98 |
| 15 | Ministral-3-8B-Instruct-2512 | 86.78 | #661 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-portuguese-nlu-sst-2-pt · How It Works · Data refreshed daily, snapshot 2026-10-20.