EuroEval Croatian NLU - Multi Wiki QA HR: leaderboard
The "Multi Wiki QA HR" subtask of EuroEval Croatian NLU. Croatian-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. 211 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Ministral-3-14B-Reasoning-2512 | 73.9 | #540 |
| 2 | Olmo-3.1-32B-Instruct-SFT | 70.12 | #879 |
| 3 | Qwen3.5-4B-Base | 69.54 | #798 |
| 4 | Bielik-11B-v2.3-Instruct | 69.45 | #693 |
| 5 | SOLAR-10.7B-v1.0 | 68.75 | #1253 |
| 6 | Mistral Small 3.2 | 68.62 | #605 |
| 7 | Qwen 3 32B (Non-reasoning) | 68.32 | #443 (Qwen 3 32B) |
| 8 | Qwen 3 14B (Non-reasoning) | 68.16 | #537 (Qwen 3 14B) |
| 9 | Ministral-3-14B-Instruct-2512 | 67.67 | #615 |
| 10 | Gemini 3 Pro (Preview) | 67.64 | #66 |
| 11 | GPT-5.2 | 67.02 | #110 |
| 12 | Llama 3.1 70B | 66.7 | #592 |
| 13 | Voxtral-Small-24B-2507 | 66.49 | #581 |
| 14 | Llama-3.1-Tulu-3-8B-SFT | 66.44 | #985 |
| 15 | Ministral-3-3B-Reasoning-2512 | 66.43 | #932 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-croatian-nlu-multi-wiki-qa-hr · How It Works · Data refreshed daily, snapshot 2026-10-20.