EuroEval Bosnian NLU - Multi Wiki QA BS: leaderboard
The "Multi Wiki QA BS" subtask of EuroEval Bosnian NLU. Bosnian-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. 211 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mistral Small 3.2 | 76.45 | #605 |
| 2 | Qwen3.5-4B-Base | 76.27 | #798 |
| 3 | Ministral-3-14B-Reasoning-2512 | 75.83 | #540 |
| 4 | Olmo-3.1-32B-Instruct-SFT | 75.21 | #879 |
| 5 | gemma-3-27B-pt | 74.64 | #502 |
| 6 | Bielik-11B-v2.3-Instruct | 74.41 | #693 |
| 7 | SOLAR-10.7B-v1.0 | 73.79 | #1253 |
| 8 | Voxtral-Small-24B-2507 | 73.56 | #581 |
| 9 | gemma-3-12B-pt | 73.53 | #686 |
| 10 | Mistral Small 3.1 | 72.46 | #610 |
| 11 | Llama 3.1 70B | 71.84 | #592 |
| 12 | Qwen 3 14B (Non-reasoning) | 71.74 | #537 (Qwen 3 14B) |
| 13 | Ministral-3-14B-Instruct-2512 | 71.17 | #615 |
| 14 | Qwen 3 32B (Non-reasoning) | 70.93 | #443 (Qwen 3 32B) |
| 15 | Ministral-3-8B-Instruct-2512 | 70.31 | #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-bosnian-nlu-multi-wiki-qa-bs · How It Works · Data refreshed daily, snapshot 2026-10-20.