EuroEval Bulgarian NLU - MultiWikiQA BG: leaderboard
The "MultiWikiQA BG" subtask of EuroEval Bulgarian NLU. Bulgarian-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. 212 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Ministral-3-14B-Reasoning-2512 | 80.66 | #540 |
| 2 | Mistral Small 3.2 | 79.74 | #605 |
| 3 | SOLAR-10.7B-v1.0 | 79.17 | #1253 |
| 4 | gemma-3-27B-pt | 78.42 | #502 |
| 5 | gemma-3-12B-pt | 77.91 | #686 |
| 6 | Qwen3.5-4B-Base | 77.91 | #798 |
| 7 | Llama 3.1 70B | 77.58 | #592 |
| 8 | Voxtral-Small-24B-2507 | 77.18 | #581 |
| 9 | Qwen 3 14B (Non-reasoning) | 77.08 | #537 (Qwen 3 14B) |
| 10 | Mistral Small 3.1 | 76.8 | #610 |
| 11 | Ministral-3-14B-Instruct-2512 | 75.41 | #615 |
| 12 | tiny-aya-water | 74.66 | |
| 13 | Qwen 3 32B (Non-reasoning) | 74.6 | #443 (Qwen 3 32B) |
| 14 | Llama 3.1 8B | 74.41 | #1151 |
| 15 | Tiny Aya Global | 73.92 | #1325 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-bulgarian-nlu-multiwikiqa-bg · How It Works · Data refreshed daily, snapshot 2026-10-20.