EuroEval Portuguese NLU - MultiWikiQA PT: leaderboard
The "MultiWikiQA 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: Character F1 (%). Source: euroeval.com. Saturation forecast: Not forecast. 438 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Bielik-11B-v2.3-Instruct | 80.57 | #693 |
| 2 | SOLAR-10.7B-v1.0 | 79.82 | #1253 |
| 3 | Ministral-3-14B-Reasoning-2512 | 79.73 | #540 |
| 4 | Qwen 3 32B (Non-reasoning) | 79.03 | #443 (Qwen 3 32B) |
| 5 | Llama 3.1 70B | 78.69 | #592 |
| 6 | Llama 3 70B | 78.57 | #631 |
| 7 | Qwen 3 14B (Non-reasoning) | 78.55 | #537 (Qwen 3 14B) |
| 8 | Gemma 2 27B | 78.36 | #806 |
| 9 | Falcon3-7B-Instruct | 78.3 | #1305 |
| 10 | Magistral Small 1.2 | 78.22 | #570 |
| 11 | Llama 2 70B Base | 78.11 | #1025 |
| 12 | Yi 1.5 34B | 77.91 | #807 |
| 13 | Gemma 3n E4B | 77.48 | #878 |
| 14 | Mistral Small 3.1 | 77.38 | #610 |
| 15 | Mixtral 8x7B (v0.1) | 77.35 | #1083 |
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-multiwikiqa-pt · How It Works · Data refreshed daily, snapshot 2026-10-20.