EuroEval Slovene NLU - MultiWikiQA SL: leaderboard
The "MultiWikiQA SL" subtask of EuroEval Slovene NLU. Slovene-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: Around January 2027. 204 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mistral Small 3.2 | 76.44 | #605 |
| 2 | Llama 3.1 70B | 74.14 | #592 |
| 3 | Mistral Small 3.1 | 73.22 | #610 |
| 4 | gemma-3-27B-pt | 72.95 | #502 |
| 5 | Ministral-3-14B-Reasoning-2512 | 72.64 | #540 |
| 6 | Qwen 3 14B (Non-reasoning) | 72.27 | #537 (Qwen 3 14B) |
| 7 | SOLAR-10.7B-v1.0 | 72.15 | #1253 |
| 8 | Qwen 3 32B (Non-reasoning) | 71.61 | #443 (Qwen 3 32B) |
| 9 | Voxtral-Small-24B-2507 | 71.36 | #581 |
| 10 | Granite 4.0 H Small | 70.68 | #665 |
| 11 | Llama-3.1-Tulu-3-8B-SFT | 70.16 | #985 |
| 12 | Bielik-11B-v2.3-Instruct | 69.3 | #693 |
| 13 | Qwen 3 8B (Non-reasoning) | 68.43 | #674 (Qwen 3 8B) |
| 14 | Qwen3.5-4B-Base | 68.33 | #798 |
| 15 | Qwen 3 4B 2507 (Thinking) | 68.17 | #538 (Qwen 3 4B 2507) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-slovene-nlu-multiwikiqa-sl · How It Works · Data refreshed daily, snapshot 2026-10-20.