EuroEval Latvian NLU - MultiWikiQA LV: leaderboard
The "MultiWikiQA LV" subtask of EuroEval Latvian NLU. Latvian-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. 233 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | gemma-3-27B-pt | 73.01 | #502 |
| 2 | Mistral Small 3.2 | 72.5 | #605 |
| 3 | Qwen 3 14B (Non-reasoning) | 72.18 | #537 (Qwen 3 14B) |
| 4 | Llama 3.1 70B | 72.11 | #592 |
| 5 | Magistral Small 1.2 | 71.87 | #570 |
| 6 | Ministral-3-14B-Reasoning-2512 | 70.52 | #540 |
| 7 | Qwen3.5-4B-Base | 69.86 | #798 |
| 8 | Qwen 3 32B (Non-reasoning) | 69.85 | #443 (Qwen 3 32B) |
| 9 | Mistral Small 3.1 | 69.2 | #610 |
| 10 | Voxtral-Small-24B-2507 | 68.76 | #581 |
| 11 | Ministral-3-3B-Reasoning-2512 | 68.05 | #932 |
| 12 | Qwen 2.5 72B Instruct | 68.04 | #454 |
| 13 | Qwen 3 14B | 67.91 | #537 |
| 14 | cogito-v1-preview-qwen-14B | 67.49 | |
| 15 | gemma-3-12B-pt | 67.15 | #686 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=euroeval-latvian-nlu-multiwikiqa-lv · How It Works · Data refreshed daily, snapshot 2026-10-20.